This document proposes algorithms using geometric functions like spirals and Lissajous curves to generate smoother trajectories for UAVs performing area coverage compared to the traditional lawnmower approach. The smoother trajectories are expected to reduce the time and energy required for area coverage by 22% and 30% respectively while still achieving at least 98% coverage. The algorithms will be implemented using ROS and experiments conducted in Gazebo simulation to evaluate the performance of the proposed methods versus the lawnmower approach.
The document discusses using optimized neural networks for short-term wind speed forecasting. It proposes using parametric recurrent neural networks (PRNNs) with an improved activation function that includes a logarithmic parameter "p" to optimize the network size. The PRNNs are trained to predict wind speed using historical wind farm data. Simulation results show the PRNNs more accurately predict wind speed up to 180 minutes in the future compared to numerical methods using polynomials. The value of the "p" parameter can identify linearly dependent neurons that can be combined to reduce the optimized network size.
Mars CubeSat Telecom Relay Constellation_JPL FinalRohan Deshmukh
The document proposes a 4-CubeSat constellation in Mars orbit to augment telecommunications relay capabilities for surface assets. The CubeSats would be deployed from a 2022 Mars Telecom Orbiter into a 350 km circular orbit with 70 degree inclination. Each 6U CubeSat would carry an IRIS transponder, Ka-band reflectarray antenna, and UHF antennas to provide direct-to/from-Earth relay and cross-link relay with orbiters. Analysis showed the design could meet power and propulsion requirements to operate for two Mars years.
The SAGAN mission proposes using a constellation of four microsatellites in a tetrahedral formation to study the impact of atmospheric waves, neutral forcing, and ionospheric currents on low and mid-latitude plasma structuring. The satellites would carry instruments to measure electric and magnetic fields, electron density, and ion composition. Maintaining the tetrahedral formation over multiple orbital planes would allow for spatial correlation measurements to understand plasma dynamics in the ionosphere. The proposed one to two year mission aims to advance understanding of ionospheric coupling processes.
Digital Heritage Documentation Via TLS And Photogrammetry Case Studytheijes
In the last decade, several manual tradition measurement techniques were used to document the heritage buildings around the word; however, some of these techniques take a long time, often lack completeness, and may sometimes give unreliable information. In contrast, terrestrial laser scanning “TLS” surveys and Photogrammetry have already been undertaken in several heritage sites in the United Kingdom and other countries of Europe as a new method of documenting heritagesites. This paper focuses on using the TLS and Photogrammetry methods to document one of the important houses in Historic Jeddah, Saudi Arabia, which is Nasif Historical House, as an example of Digital Heritage Documentation (DHD).
Nonlinear filtering approaches to field mapping by sampling using mobile sensorsijassn
This work proposes a novel application of existing powerful nonlinear filters, such as the standard
Extended Kalman Filter (EKF), some of its variants and the standard Unscented Kalman Filter (UKF), to
the estimation of a continuous spatio-temporal field that is spread over a wide area, and hence represented
by a large number of parameters when parameterized. We couple these filters with the powerful scheme of
adaptive sampling performed by a single mobile sensor, and investigate their performances with a view to
significantly improving the speed and accuracy of the overall field estimation. An extensive simulation work
was carried out to show that different variants of the standard EKF and the standard UKF can be used to
improve the accuracy of the field estimate. This paper also aims to provide some guideline for the user of
these filters in reaching a practical trade-off between the desired field estimation accuracy and the
required computational load.
This document discusses using variational inference methods to perform online change point detection when the underlying model is not in the exponential family. Specifically:
- It develops variational approximations to the posterior on change point times for efficient inference when the underlying model, like a Rice distribution, does not have tractable posterior predictive distributions.
- It applies this methodology to radar tracking, using a Rice distribution to model signal-to-noise ratio features, and develops a variational method for inferring the parameters of the non-exponential Rice distribution online.
- It provides improvements to online variational inference that allow updating approximations efficiently as new data arrives, enabling the use of variational inference within the Bayesian online change point detection framework.
Analysis of Time Diversity Gain for Satellite Communication Link based on Ku-...IJECEIAES
This paper reports a study on mitigation of propagation impairments on Earth–space communication links. The study uses time diversity as a technique for mitigating rain propagation impairment in order to rectify rain fade. Rain attenuation time series along earth-to-satellite link were measured for two years period at 12.255 GHz in Malaysia. The time diversity technique was applied on measured rain fade to investigate the level of possible improvement in system. Time diversity gain from measured oneminute rain attenuation for two years period was estimated and significant improvement was observed with different delays of time. These findings will be utilized as a useful tool for link designers to apply time diversity as a rain fade mitigation technique in Earth-satellite communications systems.
The document describes a proposed concept for a constellation of 4 CubeSats in Mars orbit to augment communication relay capabilities for assets on the Martian surface and in orbit. Key aspects of the concept include using the CubeSats to relay data between landed assets and orbiters using UHF bands, and between the CubeSats and Earth using Ka-band. Mass and power budgets have been analyzed to demonstrate the feasibility of the concept. Future work areas are identified such as further analyzing data volume, thermal, trajectory, and cost aspects.
The document discusses using optimized neural networks for short-term wind speed forecasting. It proposes using parametric recurrent neural networks (PRNNs) with an improved activation function that includes a logarithmic parameter "p" to optimize the network size. The PRNNs are trained to predict wind speed using historical wind farm data. Simulation results show the PRNNs more accurately predict wind speed up to 180 minutes in the future compared to numerical methods using polynomials. The value of the "p" parameter can identify linearly dependent neurons that can be combined to reduce the optimized network size.
Mars CubeSat Telecom Relay Constellation_JPL FinalRohan Deshmukh
The document proposes a 4-CubeSat constellation in Mars orbit to augment telecommunications relay capabilities for surface assets. The CubeSats would be deployed from a 2022 Mars Telecom Orbiter into a 350 km circular orbit with 70 degree inclination. Each 6U CubeSat would carry an IRIS transponder, Ka-band reflectarray antenna, and UHF antennas to provide direct-to/from-Earth relay and cross-link relay with orbiters. Analysis showed the design could meet power and propulsion requirements to operate for two Mars years.
The SAGAN mission proposes using a constellation of four microsatellites in a tetrahedral formation to study the impact of atmospheric waves, neutral forcing, and ionospheric currents on low and mid-latitude plasma structuring. The satellites would carry instruments to measure electric and magnetic fields, electron density, and ion composition. Maintaining the tetrahedral formation over multiple orbital planes would allow for spatial correlation measurements to understand plasma dynamics in the ionosphere. The proposed one to two year mission aims to advance understanding of ionospheric coupling processes.
Digital Heritage Documentation Via TLS And Photogrammetry Case Studytheijes
In the last decade, several manual tradition measurement techniques were used to document the heritage buildings around the word; however, some of these techniques take a long time, often lack completeness, and may sometimes give unreliable information. In contrast, terrestrial laser scanning “TLS” surveys and Photogrammetry have already been undertaken in several heritage sites in the United Kingdom and other countries of Europe as a new method of documenting heritagesites. This paper focuses on using the TLS and Photogrammetry methods to document one of the important houses in Historic Jeddah, Saudi Arabia, which is Nasif Historical House, as an example of Digital Heritage Documentation (DHD).
Nonlinear filtering approaches to field mapping by sampling using mobile sensorsijassn
This work proposes a novel application of existing powerful nonlinear filters, such as the standard
Extended Kalman Filter (EKF), some of its variants and the standard Unscented Kalman Filter (UKF), to
the estimation of a continuous spatio-temporal field that is spread over a wide area, and hence represented
by a large number of parameters when parameterized. We couple these filters with the powerful scheme of
adaptive sampling performed by a single mobile sensor, and investigate their performances with a view to
significantly improving the speed and accuracy of the overall field estimation. An extensive simulation work
was carried out to show that different variants of the standard EKF and the standard UKF can be used to
improve the accuracy of the field estimate. This paper also aims to provide some guideline for the user of
these filters in reaching a practical trade-off between the desired field estimation accuracy and the
required computational load.
This document discusses using variational inference methods to perform online change point detection when the underlying model is not in the exponential family. Specifically:
- It develops variational approximations to the posterior on change point times for efficient inference when the underlying model, like a Rice distribution, does not have tractable posterior predictive distributions.
- It applies this methodology to radar tracking, using a Rice distribution to model signal-to-noise ratio features, and develops a variational method for inferring the parameters of the non-exponential Rice distribution online.
- It provides improvements to online variational inference that allow updating approximations efficiently as new data arrives, enabling the use of variational inference within the Bayesian online change point detection framework.
Analysis of Time Diversity Gain for Satellite Communication Link based on Ku-...IJECEIAES
This paper reports a study on mitigation of propagation impairments on Earth–space communication links. The study uses time diversity as a technique for mitigating rain propagation impairment in order to rectify rain fade. Rain attenuation time series along earth-to-satellite link were measured for two years period at 12.255 GHz in Malaysia. The time diversity technique was applied on measured rain fade to investigate the level of possible improvement in system. Time diversity gain from measured oneminute rain attenuation for two years period was estimated and significant improvement was observed with different delays of time. These findings will be utilized as a useful tool for link designers to apply time diversity as a rain fade mitigation technique in Earth-satellite communications systems.
The document describes a proposed concept for a constellation of 4 CubeSats in Mars orbit to augment communication relay capabilities for assets on the Martian surface and in orbit. Key aspects of the concept include using the CubeSats to relay data between landed assets and orbiters using UHF bands, and between the CubeSats and Earth using Ka-band. Mass and power budgets have been analyzed to demonstrate the feasibility of the concept. Future work areas are identified such as further analyzing data volume, thermal, trajectory, and cost aspects.
Airport Runway Detection Based On ANN AlgorithmIJTET Journal
This document presents a methodology for detecting airport runways in satellite images. The methodology consists of two main stages: 1) a segmentation stage that classifies image regions as "probably runway" or "not runway" based on textural features, and 2) a shape detection stage that analyzes segmented regions to identify long parallel lines characteristic of runway sides. The document describes various textural features extracted for segmentation, including Haralick textures, Gabor filters, Fourier spectra, and wavelet analyses. It also details an artificial neural network classifier trained on these features to perform the segmentation. Experimental results demonstrate the algorithm can successfully detect runways in satellite imagery.
IJERD (www.ijerd.com) International Journal of Engineering Research and Devel...IJERD Editor
This document proposes a modified design for a three-loop lateral missile autopilot based on linear quadratic regulator (LQR) and a reduced order observer. The design improves on previous three-loop autopilot designs by 1) reducing the initial negative peak in response due to non-minimum phase properties, 2) reaching a steady state flight path rate of 1 for a unit step input command, eliminating steady state error. The modified design uses a reduced order Das & Ghosal observer, output feedback, and LQR state feedback to optimally place closed loop poles for desired time domain performance. Simulation results demonstrate the improved performance over classical two-loop and three-loop autopilot designs.
This document proposes using cloud computing to rapidly generate earthquake shake maps through stochastic simulations. It describes implementing a map-reduce model on the Microsoft Azure cloud to parallelize simulations for many receivers after an earthquake. Results like peak ground acceleration maps could be produced within minutes to aid scientists and emergency responders. The method was tested in Greece and provides a cost-effective way to process simulations as needed without dedicated computing resources.
CLIC Accelerator: status, plans and outlook asafrona
This document provides an overview of the status, plans, and outlook for the Compact Linear Collider (CLIC) accelerator project. Key points include:
1) CLIC has made progress optimizing its design for lower initial energies around 380 GeV based on LHC results, with the goal of reducing costs and power requirements for early stages.
2) Recent test facility results like those at CTF3 have demonstrated drive beam generation and two-beam acceleration, advancing critical technologies.
3) The collaboration aims to provide a staged implementation plan up to 3 TeV by 2019 to inform the next European strategy update, accounting for further LHC data and optimizing costs.
4) Advancing X-
02 Modelling strategies for Nuclear Probabilistic Safety Assessment in case o...RCCSRENKEI
This document discusses modelling strategies for probabilistic safety assessment of nuclear sites in the case of natural external events. It provides insights into probabilistic safety assessment methodology for multi-hazard events. It also discusses ongoing developments including using model reduction techniques to create virtual charts for seismic risk assessment, and leveraging high performance computing. The goal is to improve safety assessment methods through theoretical enhancements and applying outcomes to demonstration problems.
CLIC plans for the next European strategy update, and beyondasafrona
The document discusses plans for CLIC (Compact Linear Collider) in preparation for the next European strategy update in 2019/2020. It outlines the development phase goals of updating physics studies based on LHC results, developing a project plan with costing and staging, and completing key technical R&D. It also discusses plans for the preparation phase from 2020-2025, including developing final prototypes, establishing industrial production, and building facilities for testing components. Key challenges discussed are fully developing the industrial basis and qualifying components through testing before the planned start of construction.
This document discusses using air traffic complexity metrics rather than just flight counts and sector capacities to predict air traffic controller workload and dynamically configure airspace. Previous research found relationships between relevant complexity metrics and sector status (merged, manned, or split). This work aims to use those metrics and predicted sector statuses to build realistic airspace configuration schedules on a minute-by-minute basis, and compare them to actual configurations. The goals are more accurate workload assessment and improved airspace configuration over current pre-defined schedules.
MACHINE LEARNING FOR SATELLITE-GUIDED WATER QUALITY MONITORINGVisionGEOMATIQUE2014
The document discusses using machine learning techniques for satellite-guided water quality monitoring. It covers using machine learning algorithms to automatically develop empirical models from multimodal satellite and field data sets. Machine learning can help construct nonlinear mappings between satellite measurements and water quality products and optimize in-situ data collection through mission planning. Experimental results are shown applying these techniques to map water quality metrics like chlorophyll-a and total suspended solids using MODIS satellite images of Lake Winnipeg.
This document discusses plans for further research and development of the Compact Linear Collider (CLIC) project. It outlines key goals for 2013-2018, which include optimizing the machine design for different energy stages, further technical developments, and system tests using facilities like CTF3. The long term goal is to present a credible staged implementation plan for CLIC to follow the LHC program by the next European strategy update in 2018. Ongoing work includes technical demonstrations and studies of the drive beam concept, high gradient accelerating structures, beam instrumentation, and superconducting magnets.
This document discusses an algorithm for estimating radar cross section (RCS). RCS is a measure of how detectable an object is by radar. The algorithm takes in data about a target like its material, size, distance from radar, and angle, and uses the radar equation to calculate the RCS. The algorithm was tested on different target types and showed varying RCS values, with aircraft generally higher than humans or insects. Calculating RCS allows analysis of radar system performance independent of specific system parameters.
This document summarizes research conducted using supercomputers to enable artificial intelligence applications for analyzing large earth science data. It discusses two major functions: designing efficient simulations and developing intelligent data mining methods. Specific projects are described, including simulating climate models on the Sunway TaihuLight, simulating earthquakes, and using remote sensing data and deep learning to map land cover more accurately, detect oil palm trees, and create more accurate urban land use maps. The research enables digital earth modeling to simulate, analyze, understand, predict, and mitigate earth science issues.
The document summarizes a web-based weather monitoring system in Palermo, Italy that collects temperature and other data from several weather stations around the city. It uses the data to analyze trends like the urban heat island effect and creates forecasts using artificial neural networks. The forecasts showed good accuracy within a 1-2 degree range when predicting temperatures a week in advance for different areas around Palermo. Future work will focus on longer-term forecasting to study thermal comfort conditions in the city.
Accès ouvert aux données météorologiques d’Environnement CanadaVisionGEOMATIQUE2014
Deux services de données géospatiales offerts par le Service météorologique Canada (Environnement Canada) seront présentés.
Le Datamart du SMC permet de télécharger des données brutes produites par le Service météorologique du Canada (SMC) et rend possible la cueillette automatisée des produits et données météorologiques du SMC. Une grande diversité de données s’y retrouvent, tels les avertissements météo, les prévisions publiques en XML, les données brutes de prévision, les observations météorologiques et bien davantage. Le Datamart du SMC jouit d’une grande popularité avec plus de 12 millions d’accès quotidiennement.
Les services web géospatiaux GeoMet permettent au public d’intégrer dans leurs outils SIG, cartes interactives ou appareil mobile, les données brutes de modèles de prévision numérique du temps et de la mosaïque nord-américaine des radars météo par le biais de deux standards de l’Open Geospatial Consortium : le WMS et le KML. Qu’il s’agisse par exemple de la pluie observée par les radar météo ou les prévisions de précipitation de neige ou une animation des prévisions de températures , les utilisations de ces données sont nombreuses et d’une grande valeur pour de nombreux décideurs.
The document discusses the Compact Linear Collider (CLIC) project. CLIC is a proposed linear collider that would collide electrons and positrons at very high energies up to 3 TeV. Its goal is to produce new heavy particles for study. The CLIC collaboration involves over 70 institutes from 29 countries. Current efforts include developing and testing key CLIC technologies like two-beam acceleration and nano-positioning, building test facilities like CTF3, and preparing for the next update to the European strategy for particle physics in 2018/19.
This document describes two distributed-memory parallelization schemes for efficiently parallelizing an explicit time-domain volume integral equation solver on the IBM Blue Gene/P supercomputer. The first scheme distributes the computationally intensive tested field computations among processors while storing the source field time histories on each processor, requiring all-to-all global communications. The second scheme distributes both the source fields and tested field computations, requiring sequential global communications. Numerical results show that both schemes scale well on Blue Gene/P, and the second more memory-efficient scheme allows solving problems with up to 3 million unknowns without acceleration. The parallel solver is demonstrated on the problem of light scattering from a red blood cell.
The document discusses the CLIC project, which aims to build a linear collider to study particle physics. It outlines the main activities from 2014-2015, which included developing the accelerator complex and testing key components. It also summarizes the physics goals of CLIC in studying the Higgs boson and searching for new particles at higher energies than the LHC. The document provides an overview of the CLIC design and layout, and outlines the development phases planned towards potential construction starting in 2024.
The document proposes developing a lunar hopper vehicle powered by a nuclear thermal rocket engine. It would take off and land repeatedly to gather soil samples from multiple locations on the moon's surface. An optimization problem is set up in MATLAB to determine the ideal dry mass and mass ratio to maximize the number of landing sites visited. The analysis finds that a design using a SNRE (Small Nuclear Reactor Engine) with a 3500 kg dry mass could achieve up to 28 sample retrievals from sites 5 km apart. A conceptual design of the vehicle is then presented, including subsystem mass budgets and dimensioned drawings. Future work areas are identified to further develop the design.
CALIFES could be used as a test facility after CTF3 stops operations in 2016. It would allow testing of various beam instrumentation, high-gradient structures, components, and systems for future projects. Maintaining a beam test capability at CERN is important for ongoing accelerator R&D. Potential experiments include diagnostics development, impedance measurements, irradiation tests, and studies of plasma wakefield acceleration. Upgrades like additional RF power could improve CALIFES' capabilities for supporting these types of beam tests beyond the lifetime of CTF3.
This document describes a new recursive Monte Carlo simulation algorithm called the Sampled Path Set Algorithm (SPSA) for modeling complex k-out-of-n reliability systems. The SPSA uses a graph representation of a reliability block diagram and recursively searches the graph to determine system response based on the system state vector at each simulation iteration, allowing modeling of systems with general component failure and repair distributions and large numbers of components. Existing methods for analyzing such systems using tie/cut sets have limitations as the number of sets grows non-linearly with increased system complexity. The SPSA provides a more efficient alternative with linear growth in processing and memory requirements.
Reliable Positioning with Hybrid Antenna Model for Aerial Wireless Sensor and...M. Ilhan Akbas
Aerial wireless sensor and actor networks are composed of multiple unmanned aerial vehicles. An actor node in the network has the capabilities of both acting on the environment and also performing networking functionalities for sensor nodes. Thus, positioning of actors is critical for the efficient data collection. In this study, we propose an actor positioning strategy, which utilizes a hybrid antenna model that combines the complimentary features of an isotropic omni radio and directional antennas. We present a distributed algorithm for fast neighbor discovery with the hybrid antenna. The omni module of the hybrid antenna is used to form a self organizing network and the directional module is used for reliable data transmission. Extensive simulations show that our protocol improves the packet reception ratio by up to 50% compared to omnidirectional antenna. Moreover, the network reorganization delay is also reduced. The tradeoff between coverage and reorganization delay is also illustrated.
Để xem full tài liệu Xin vui long liên hệ page để được hỗ trợ
:
https://www.facebook.com/garmentspace/
https://www.facebook.com/thuvienluanvan01
HOẶC
https://www.facebook.com/thuvienluanvan01
https://www.facebook.com/thuvienluanvan01
tai lieu tong hop, thu vien luan van, luan van tong hop, do an chuyen nganh
The document outlines the topics to be covered in a presentation: the positive and negative aspects of physical activity, why employees are not engaging in it, solutions to convince directors to sign up employees for a company sports program, and the long-term solution. It then discusses making excuses, stress, and not seeing physical activity as beneficial as reasons why employees do not engage. Creating an inspirational workshop and health measurements at the office are proposed as short-term solutions, while a business operating plan is suggested as the long-term solution.
Airport Runway Detection Based On ANN AlgorithmIJTET Journal
This document presents a methodology for detecting airport runways in satellite images. The methodology consists of two main stages: 1) a segmentation stage that classifies image regions as "probably runway" or "not runway" based on textural features, and 2) a shape detection stage that analyzes segmented regions to identify long parallel lines characteristic of runway sides. The document describes various textural features extracted for segmentation, including Haralick textures, Gabor filters, Fourier spectra, and wavelet analyses. It also details an artificial neural network classifier trained on these features to perform the segmentation. Experimental results demonstrate the algorithm can successfully detect runways in satellite imagery.
IJERD (www.ijerd.com) International Journal of Engineering Research and Devel...IJERD Editor
This document proposes a modified design for a three-loop lateral missile autopilot based on linear quadratic regulator (LQR) and a reduced order observer. The design improves on previous three-loop autopilot designs by 1) reducing the initial negative peak in response due to non-minimum phase properties, 2) reaching a steady state flight path rate of 1 for a unit step input command, eliminating steady state error. The modified design uses a reduced order Das & Ghosal observer, output feedback, and LQR state feedback to optimally place closed loop poles for desired time domain performance. Simulation results demonstrate the improved performance over classical two-loop and three-loop autopilot designs.
This document proposes using cloud computing to rapidly generate earthquake shake maps through stochastic simulations. It describes implementing a map-reduce model on the Microsoft Azure cloud to parallelize simulations for many receivers after an earthquake. Results like peak ground acceleration maps could be produced within minutes to aid scientists and emergency responders. The method was tested in Greece and provides a cost-effective way to process simulations as needed without dedicated computing resources.
CLIC Accelerator: status, plans and outlook asafrona
This document provides an overview of the status, plans, and outlook for the Compact Linear Collider (CLIC) accelerator project. Key points include:
1) CLIC has made progress optimizing its design for lower initial energies around 380 GeV based on LHC results, with the goal of reducing costs and power requirements for early stages.
2) Recent test facility results like those at CTF3 have demonstrated drive beam generation and two-beam acceleration, advancing critical technologies.
3) The collaboration aims to provide a staged implementation plan up to 3 TeV by 2019 to inform the next European strategy update, accounting for further LHC data and optimizing costs.
4) Advancing X-
02 Modelling strategies for Nuclear Probabilistic Safety Assessment in case o...RCCSRENKEI
This document discusses modelling strategies for probabilistic safety assessment of nuclear sites in the case of natural external events. It provides insights into probabilistic safety assessment methodology for multi-hazard events. It also discusses ongoing developments including using model reduction techniques to create virtual charts for seismic risk assessment, and leveraging high performance computing. The goal is to improve safety assessment methods through theoretical enhancements and applying outcomes to demonstration problems.
CLIC plans for the next European strategy update, and beyondasafrona
The document discusses plans for CLIC (Compact Linear Collider) in preparation for the next European strategy update in 2019/2020. It outlines the development phase goals of updating physics studies based on LHC results, developing a project plan with costing and staging, and completing key technical R&D. It also discusses plans for the preparation phase from 2020-2025, including developing final prototypes, establishing industrial production, and building facilities for testing components. Key challenges discussed are fully developing the industrial basis and qualifying components through testing before the planned start of construction.
This document discusses using air traffic complexity metrics rather than just flight counts and sector capacities to predict air traffic controller workload and dynamically configure airspace. Previous research found relationships between relevant complexity metrics and sector status (merged, manned, or split). This work aims to use those metrics and predicted sector statuses to build realistic airspace configuration schedules on a minute-by-minute basis, and compare them to actual configurations. The goals are more accurate workload assessment and improved airspace configuration over current pre-defined schedules.
MACHINE LEARNING FOR SATELLITE-GUIDED WATER QUALITY MONITORINGVisionGEOMATIQUE2014
The document discusses using machine learning techniques for satellite-guided water quality monitoring. It covers using machine learning algorithms to automatically develop empirical models from multimodal satellite and field data sets. Machine learning can help construct nonlinear mappings between satellite measurements and water quality products and optimize in-situ data collection through mission planning. Experimental results are shown applying these techniques to map water quality metrics like chlorophyll-a and total suspended solids using MODIS satellite images of Lake Winnipeg.
This document discusses plans for further research and development of the Compact Linear Collider (CLIC) project. It outlines key goals for 2013-2018, which include optimizing the machine design for different energy stages, further technical developments, and system tests using facilities like CTF3. The long term goal is to present a credible staged implementation plan for CLIC to follow the LHC program by the next European strategy update in 2018. Ongoing work includes technical demonstrations and studies of the drive beam concept, high gradient accelerating structures, beam instrumentation, and superconducting magnets.
This document discusses an algorithm for estimating radar cross section (RCS). RCS is a measure of how detectable an object is by radar. The algorithm takes in data about a target like its material, size, distance from radar, and angle, and uses the radar equation to calculate the RCS. The algorithm was tested on different target types and showed varying RCS values, with aircraft generally higher than humans or insects. Calculating RCS allows analysis of radar system performance independent of specific system parameters.
This document summarizes research conducted using supercomputers to enable artificial intelligence applications for analyzing large earth science data. It discusses two major functions: designing efficient simulations and developing intelligent data mining methods. Specific projects are described, including simulating climate models on the Sunway TaihuLight, simulating earthquakes, and using remote sensing data and deep learning to map land cover more accurately, detect oil palm trees, and create more accurate urban land use maps. The research enables digital earth modeling to simulate, analyze, understand, predict, and mitigate earth science issues.
The document summarizes a web-based weather monitoring system in Palermo, Italy that collects temperature and other data from several weather stations around the city. It uses the data to analyze trends like the urban heat island effect and creates forecasts using artificial neural networks. The forecasts showed good accuracy within a 1-2 degree range when predicting temperatures a week in advance for different areas around Palermo. Future work will focus on longer-term forecasting to study thermal comfort conditions in the city.
Accès ouvert aux données météorologiques d’Environnement CanadaVisionGEOMATIQUE2014
Deux services de données géospatiales offerts par le Service météorologique Canada (Environnement Canada) seront présentés.
Le Datamart du SMC permet de télécharger des données brutes produites par le Service météorologique du Canada (SMC) et rend possible la cueillette automatisée des produits et données météorologiques du SMC. Une grande diversité de données s’y retrouvent, tels les avertissements météo, les prévisions publiques en XML, les données brutes de prévision, les observations météorologiques et bien davantage. Le Datamart du SMC jouit d’une grande popularité avec plus de 12 millions d’accès quotidiennement.
Les services web géospatiaux GeoMet permettent au public d’intégrer dans leurs outils SIG, cartes interactives ou appareil mobile, les données brutes de modèles de prévision numérique du temps et de la mosaïque nord-américaine des radars météo par le biais de deux standards de l’Open Geospatial Consortium : le WMS et le KML. Qu’il s’agisse par exemple de la pluie observée par les radar météo ou les prévisions de précipitation de neige ou une animation des prévisions de températures , les utilisations de ces données sont nombreuses et d’une grande valeur pour de nombreux décideurs.
The document discusses the Compact Linear Collider (CLIC) project. CLIC is a proposed linear collider that would collide electrons and positrons at very high energies up to 3 TeV. Its goal is to produce new heavy particles for study. The CLIC collaboration involves over 70 institutes from 29 countries. Current efforts include developing and testing key CLIC technologies like two-beam acceleration and nano-positioning, building test facilities like CTF3, and preparing for the next update to the European strategy for particle physics in 2018/19.
This document describes two distributed-memory parallelization schemes for efficiently parallelizing an explicit time-domain volume integral equation solver on the IBM Blue Gene/P supercomputer. The first scheme distributes the computationally intensive tested field computations among processors while storing the source field time histories on each processor, requiring all-to-all global communications. The second scheme distributes both the source fields and tested field computations, requiring sequential global communications. Numerical results show that both schemes scale well on Blue Gene/P, and the second more memory-efficient scheme allows solving problems with up to 3 million unknowns without acceleration. The parallel solver is demonstrated on the problem of light scattering from a red blood cell.
The document discusses the CLIC project, which aims to build a linear collider to study particle physics. It outlines the main activities from 2014-2015, which included developing the accelerator complex and testing key components. It also summarizes the physics goals of CLIC in studying the Higgs boson and searching for new particles at higher energies than the LHC. The document provides an overview of the CLIC design and layout, and outlines the development phases planned towards potential construction starting in 2024.
The document proposes developing a lunar hopper vehicle powered by a nuclear thermal rocket engine. It would take off and land repeatedly to gather soil samples from multiple locations on the moon's surface. An optimization problem is set up in MATLAB to determine the ideal dry mass and mass ratio to maximize the number of landing sites visited. The analysis finds that a design using a SNRE (Small Nuclear Reactor Engine) with a 3500 kg dry mass could achieve up to 28 sample retrievals from sites 5 km apart. A conceptual design of the vehicle is then presented, including subsystem mass budgets and dimensioned drawings. Future work areas are identified to further develop the design.
CALIFES could be used as a test facility after CTF3 stops operations in 2016. It would allow testing of various beam instrumentation, high-gradient structures, components, and systems for future projects. Maintaining a beam test capability at CERN is important for ongoing accelerator R&D. Potential experiments include diagnostics development, impedance measurements, irradiation tests, and studies of plasma wakefield acceleration. Upgrades like additional RF power could improve CALIFES' capabilities for supporting these types of beam tests beyond the lifetime of CTF3.
This document describes a new recursive Monte Carlo simulation algorithm called the Sampled Path Set Algorithm (SPSA) for modeling complex k-out-of-n reliability systems. The SPSA uses a graph representation of a reliability block diagram and recursively searches the graph to determine system response based on the system state vector at each simulation iteration, allowing modeling of systems with general component failure and repair distributions and large numbers of components. Existing methods for analyzing such systems using tie/cut sets have limitations as the number of sets grows non-linearly with increased system complexity. The SPSA provides a more efficient alternative with linear growth in processing and memory requirements.
Reliable Positioning with Hybrid Antenna Model for Aerial Wireless Sensor and...M. Ilhan Akbas
Aerial wireless sensor and actor networks are composed of multiple unmanned aerial vehicles. An actor node in the network has the capabilities of both acting on the environment and also performing networking functionalities for sensor nodes. Thus, positioning of actors is critical for the efficient data collection. In this study, we propose an actor positioning strategy, which utilizes a hybrid antenna model that combines the complimentary features of an isotropic omni radio and directional antennas. We present a distributed algorithm for fast neighbor discovery with the hybrid antenna. The omni module of the hybrid antenna is used to form a self organizing network and the directional module is used for reliable data transmission. Extensive simulations show that our protocol improves the packet reception ratio by up to 50% compared to omnidirectional antenna. Moreover, the network reorganization delay is also reduced. The tradeoff between coverage and reorganization delay is also illustrated.
Để xem full tài liệu Xin vui long liên hệ page để được hỗ trợ
:
https://www.facebook.com/garmentspace/
https://www.facebook.com/thuvienluanvan01
HOẶC
https://www.facebook.com/thuvienluanvan01
https://www.facebook.com/thuvienluanvan01
tai lieu tong hop, thu vien luan van, luan van tong hop, do an chuyen nganh
The document outlines the topics to be covered in a presentation: the positive and negative aspects of physical activity, why employees are not engaging in it, solutions to convince directors to sign up employees for a company sports program, and the long-term solution. It then discusses making excuses, stress, and not seeing physical activity as beneficial as reasons why employees do not engage. Creating an inspirational workshop and health measurements at the office are proposed as short-term solutions, while a business operating plan is suggested as the long-term solution.
Este documento fornece informações sobre a eficiência energética em sistemas de ar comprimido. Ele descreve os principais componentes de sistemas de ar comprimido, sugere etapas para identificar oportunidades de melhoria e discute áreas-chave para aumentar a eficiência, como geração, distribuição e uso final do ar comprimido. O documento também aborda aspectos econômicos e fornece referências adicionais.
There are conversations that every marriage needs to have. In this presentation, we give each conversation a location to understand the topic of discussion.
Để xem full tài liệu Xin vui long liên hệ page để được hỗ trợ
:
https://www.facebook.com/garmentspace/
https://www.facebook.com/thuvienluanvan01
HOẶC
https://www.facebook.com/thuvienluanvan01
https://www.facebook.com/thuvienluanvan01
tai lieu tong hop, thu vien luan van, luan van tong hop, do an chuyen nganh
Everyone wants better days in their marriage, but some question if those days will ever come. The truth is the hard times help make your marriage better.
Take a look at how to make it through the hard times.
RevolutionFibres Technology Strategy Project Report by DMDeveshoan Michael
RevolutionFibres is an Auckland-based company that produces nanofibres on an industrial scale using their proprietary sonic electrospinning technology. This project aims to identify lucrative areas within the specialty films market for RevolutionFibres to develop technology strategies. Initial research was conducted on the nanofibre and specialty films industries. Piezoelectric generators, oxygen barriers, fuel cells, sensors, dye sensitized solar cells, and supercapacitors were identified as industries where nanofibres could provide advantages. After analysis, supercapacitors were determined to best align with RevolutionFibres' strategy. A technology strategy canvas was developed to assist RevolutionFibres in entering the supercapacitor market within 3-5 years by brid
Giá 10k/ 5 lượt tải liên hệ page để mua https://www.facebook.com/garmentspace
Chỉ với 10k THẺ CÀO VIETTEL bạn có ngay 5 lượt download tài liệu bất kỳ do Garment Space upload, hoặc với 100k THẺ CÀO VIETTEL bạn được truy cập kho tài liệu chuyên ngành vô cùng phong phú
Liên hệ: www.facebook.com/garmentspace
En este documento se puede apreciar a detalle las ventajas y desventajas de obtener un credito agropecuario, así como tambien las maneras de conseguirlo o tramitarlos.
1) The document discusses parsing methods for context-free grammars including top-down and bottom-up approaches. Top-down parsing starts with the start symbol and works towards the leaves, while bottom-up parsing begins at the leaves and works towards the root.
2) Key aspects of parsing covered include left recursion elimination, left factoring, shift-reduce parsing which uses a stack and parsing table, and constructing parse trees from the parsing process.
3) The output of parsers can include parse sequences, parse trees, and abstract syntax trees which abstract away implementation details.
This document presents a framework that uses a modified particle swarm optimization (PSO) algorithm for unmanned aerial vehicle (UAV) path planning to support wireless network user rate requirements. The framework formulates the problem to jointly optimize the path and energy consumption while satisfying the user's sum rate constraint. It first uses line of sight probability to find an optimal destination where the UAV can provide the required downlink rate to the user. Then, the modified PSO algorithm finds the most energy-efficient path from the source to that destination. Experiments show the framework provides an obstacle-avoiding 3D path, minimizes energy use and travel time, and improves user rate over other methods in different scenarios.
Palm-sized quadrotor source localization using modified bio-inspired algorit...IJECEIAES
This paper uses a palm-sized quadrotor to demonstrate the gas source localization (GSL) in obstacle regions. The chemical substance performs a dilute dispersion in the air, and it is challenging to locate the source position. The difficulty level of the GSL increases if the location is in the obstacle region. To estimate the gas source position, it indeed uses high computational cost systems. However, it leads to a new problem: a palmsized quadrotor has a limited size and payload capacity. In this paper, a search based on the silkworm moth is employed. The algorithm works straightforwardly as the detection gas triggers its motion. A repulsion function is integrated into a searching method to compensate the algorithm for working in single surface obstacle regions. The validation data and the palm-sized quadrotor design are provided to convince the search effectiveness performance.
Advanced UAV Trajectory Generation Planning And GuidanceStephen Faucher
The document discusses trajectory generation and guidance capabilities for unmanned aerial vehicles (UAVs). It presents the Trajectory Generation and Guidance Module (TG2M), a software tool that can generate complex 6-degree-of-freedom trajectories in 3D space while meeting velocity, acceleration, orientation and time constraints. The TG2M is an extension of the Aerial Vehicle Control Language (AVCL) architecture. It uses analytical functions and polynomial interpolation to model trajectories and constraints such as kinematics. Simulation results using a helicopter simulator demonstrate the TG2M's ability to fly aggressive and simple maneuvers.
This document discusses optimizing UAV systems for rapidly surveying and reconstructing large cultural heritage sites through 3D modeling. It proposes using a system of fixed-wing, foldable, and multi-rotor drones carrying high-resolution cameras. Images would be transmitted via LTE to a base station and cloud for real-time processing into orthophotos and 3D point clouds using structure from motion. The goal is efficient, high-resolution documentation of extensive heritage areas.
This document proposes a method for optimizing task distribution in target tracking using unmanned aerial vehicles (UAVs) and edge computing. It describes a system where UAVs track targets and offload computationally intensive tasks to edge nodes for processing. The authors develop an algorithm that considers transmission energy consumption and execution time to select optimal edge nodes. Simulation results show their proposed UAV-enabled edge computing approach outperforms existing techniques in target tracking while minimizing costs.
This paper proposes a method to calculate a flight cost of an unmanned aerial vehicle (UAV) considering its change of heading angle though there are many reasons that cause the energy consumption. The proposed approach demonstrates that when a UAV moves from a starting position/point to a target/goal position/point, if the number of obstacle increases, the number of heading change would also increase. As a result, it raises the energy consumption of the UAV. It also shows that the magnitude of heading change would affect the energy consumption proportionally. The theoretical analysis as well as the simulation outcome proves the usefulness of the proposed technique.
This document summarizes a computational analysis of flow behavior over a multi-stage launch vehicle with strap-on boosters. A structured grid and commercial CFD software Fluent were used to model the 2D flow field. Finer meshes were generated near vehicle surfaces to capture shocks accurately. Both Euler and Navier-Stokes solvers were tested using various turbulence models. Results aimed to understand complex flow interactions over the multi-component vehicle configuration.
Digital Heritage Documentation Via TLS And Photogrammetry Case Studytheijes
In the last decade, several manual tradition measurement techniques were used to document the heritage buildings around the word; however, some of these techniques take a long time, often lack completeness, and may sometimes give unreliable information. In contrast, terrestrial laser scanning “TLS” surveys and Photogrammetry have already been undertaken in several heritage sites in the United Kingdom and other countries of Europe as a new method of documenting heritagesites. This paper focuses on using the TLS and Photogrammetry methods to document one of the important houses in Historic Jeddah, Saudi Arabia, which is Nasif Historical House, as an example of Digital Heritage Documentation (DHD)
Task Scheduling using Hybrid Algorithm in Cloud Computing Environmentsiosrjce
The document summarizes a proposed hybrid task scheduling algorithm called PSOCS that combines particle swarm optimization (PSO) and cuckoo search (CS) for scheduling tasks in cloud computing environments. The PSOCS algorithm aims to minimize task completion time (makespan) and improve resource utilization. It was tested in a simulation using CloudSim and showed reductions in makespan and increases in utilization compared to PSO and random scheduling algorithms.
This document summarizes a research paper that proposes a hybrid task scheduling algorithm for cloud computing environments called PSOCS. PSOCS combines the Particle Swarm Optimization (PSO) algorithm and Cuckoo Search (CS) algorithm to optimize task scheduling and minimize completion time while increasing resource utilization. The paper describes PSO and CS algorithms individually, then defines the proposed PSOCS algorithm. It evaluates PSOCS using a simulation and finds it reduces makespan and increases utilization compared to PSO and random allocation algorithms.
Hybrid of Ant Colony Optimization and Gravitational Emulation Based Load Bala...IRJET Journal
This document proposes a hybrid Ant Colony Optimization (ACO) and Gravitational Emulation Local Search (GELS) algorithm for load balancing in cloud computing. ACO is combined with GELS to take advantage of both algorithms - ACO is good for global search using pheromone trails while GELS is powerful for local search based on gravitational attraction. The hybrid algorithm is tested using CloudSim and shows improvements over existing algorithms like GA-GELS in metrics like resource utilization, makespan, and load balancing level.
1. The document presents a simulation software developed to evaluate the control system for an autonomous unmanned helicopter.
2. The simulation models the helicopter dynamics, sensors, and an extended Kalman filter. It accounts for forces like gravity, rotors, wind, and allows tuning the helicopter servos.
3. The goal is to design the guidance system without risking damage to real equipment by testing in simulation first.
This document summarizes an experiment that evaluated the efficiency of three search strategies for autonomous rendezvous in space: random, semi-autonomous, and autonomous. The experiment used LEGO robots to simulate a space tug searching for a target in a two-dimensional space, and measured the time and energy required for each strategy. It was found that the semi-autonomous strategy was most energy efficient but most time-consuming, while the autonomous strategy proved most suitable for space applications by being both reasonably energy efficient and faster. The results provide insight into optimal search algorithms for an actual space tug vehicle to implement during rendezvous and docking.
Unmanned aerial vehicles (UAVs) have potential for environmental mapping applications. This study tested a hexacopter UAV equipped with a calibrated digital camera to acquire aerial images of simulation models, including a coastal area and archaeological site. Digital photos were processed to generate orthophotos and digital elevation models. The hexacopter was able to autonomously capture images with sufficient accuracy. The results demonstrate that UAVs are a low-cost alternative to manned aircraft for small-area environmental mapping projects with limited budgets and timeframes. Future work involves using UAVs to map real coastal and archaeological sites.
Location Tracking and Smooth Path Providing SystemIRJET Journal
This document describes a location tracking and route planning system that was created to address issues with GPS reliability and provide accurate travel routes. It uses machine learning and data science algorithms like Dijkstra's algorithm and Kalman filtering. The system takes a source and destination address from the user, processes the coordinates using graph networks and algorithms to find multiple routes and the shortest route. Kalman filtering is then applied to smooth the shortest route. The optimized route is plotted on a map along with nearby amenities, weather data, and travel time. The system aims to help traffic management and provide accurate paths to reduce travel efforts. It was implemented using Python libraries and APIs to retrieve location data, weather, and map features.
Comparative Study of Indoor Navigation Systems for Autonomous FlightTELKOMNIKA JOURNAL
Recently, Unmanned Aerial Vehicles (UAVs) have attracted the society and researchers due to
the capability to perform in economic, scientific and emergency scenarios, and are being employed in large
number of applications especially during the hostile environments. They can operate autonomously for
both indoor and outdoor applications mainly including search and rescue, manufacturing, forest fire
tracking, remote sensing etc. For both environments, precise localization plays a critical role in order to
achieve high performance flight and interacting with the surrounding objects. However, for indoor areas
with degraded or denied Global Navigation Satellite System (GNSS) situation, it becomes challenging to
control UAV autonomously especially where obstacles are unidentified. A large number of techniques by
using various technologies are proposed to get rid of these limits. This paper provides a comparison of
such existing solutions and technologies available for this purpose with their strengths and limitations.
Further, a summary of current research status with unresolved issues and opportunities is provided that
would provide research directions to the researchers of the similar interests.
AUTOMATIC IDENTIFICATION OF CLOUD COVER REGIONS USING SURF ijcseit
Weather forecasting has become an indispensable application to predict the state of the atmosphere for a
future time based on cloud cover identification. But it generally needs the experience of a well-trained
meteorologist. In this paper, a novel method is proposed for automatic cloud cover estimation, typical to
Indian Territory Speeded Up Robust Feature Transform(SURF) is applied on the satellite images to obtain
the affine corrected images. The extracted cloud regions from the affine corrected images based on Otsu
threshold are superimposed on the artistic grids representing latitude and longitude over India. The
segmented cloud and grid composition drive a look up table mechanism to identify the cloud cover regions.
Owing to its simplicity, the proposed method processes the test images faster and provides accurate
segmentation for cloud cover regions.
This document discusses a study on assessing the quality of geospatial data obtained from unmanned aerial vehicles (UAVs). The study aimed to generate accurate 3D geospatial data from UAV images of a 5.26 km2 area near Banaras Hindu University in India. Two software programs, ArcGIS Pro and Pix4Dmapper, were used to process over 135 images taken with a DJI Mavic Pro drone. The positional and vertical accuracies of the UAV data were determined by calculating the root mean square error of checkpoint locations. The analysis found horizontal and vertical errors within an acceptable range, demonstrating that low-cost UAVs can be used to obtain accurate 3D spatial data for large-scale mapping
Triton UAS Technical Design Paper 2020-2021KennyPham19
The document summarizes the design of an unmanned aerial system called the Swallow created by Triton Unmanned Aerial Systems for the 2021 AUVSI student competition. Key aspects of the design include the fixed-wing aircraft chosen, the autopilot and ground control system used, computer vision systems for object detection and classification from images, and algorithms for autonomous path planning that avoids obstacles while completing mission tasks. The team's goal is to have a system that can autonomously identify and classify ground targets from images taken during flight.
Performance evaluation of a discovery and scheduling protocol for multihop ad...Minor33
This document summarizes a research paper that proposes a protocol called DICHOTOMY for integrated resource discovery and scheduling in mobile ad hoc grids. The protocol allows computational tasks to be distributed among the most resourceful nodes in a mobile ad hoc network to efficiently balance workload. Evaluation of the protocol shows that it performs proper scheduling, maintains acceptable discovery efficiency in mobility scenarios, and scales well with increasing nodes.
Similar to RDC-2016-ST-paper-final-Mukherjee.pdf (20)
Performance evaluation of a discovery and scheduling protocol for multihop ad...
RDC-2016-ST-paper-final-Mukherjee.pdf
1. Cooperative Trajectory Planning Of UAVs For
Area Coverage
Poulastya Mukherjee, Dr. Marek Schikora, Prof.Dr. Paul.G.Ploeger
Bonn-Rhein-Sieg University of Applied Sciences,
Grantham-Allee 20 53757 Sankt Augustin, Germany
Abstract. Traditional approach to coverage planning has always fo-
cused on developing new techniques for area decomposition and apply-
ing search based techniques to navigate an Unmanned Aerial Vehicle
(UAV) through the decomposed areas. A key feature ignored in these
approaches is the nature of the motion followed by the UAV. Usually
the actual motion path of an UAV, using these traditional techniques,
is characterized by sharp turns and quite a few start-stop motions. This
results in higher time requirement for completing coverage and deple-
tion of excess battery power (as more work needs to be done to move
the UAV). Considering the fact that UAVs have limited computational
capability and power supply, these two factors are equally important as
the percentage of area covered.
In this work we propose algorithms based on geometric functions which
have smoother trajectory compared to the lawnmower approach. By con-
ducting experiments using our proposed methods, we were able to prove
that algorithms with smoother trajectories take 22% lesser time and
consume 30% lesser energy compared to lawnmower approach - while
achieving atleast 98% coverage.
Keywords: UAV, Geometry Based Trajectories, Area Coverage, Lawn-
mower Approach
1 Introduction
The late 20th century marked the beginning of a shift from electronics age to in-
formation age. Be it health-care, entertainment, defense, industries, having thor-
ough information about every aspect of operation is considered critical. Having
requisite amount of information not only helps in devising detailed plans but
also provides insights about how to deal with a situation in which a plan has
failed.
For situations like post disaster damage mitigation, formulating defense recon-
naissance plans, or conducting geological survey of a terrain, gathering infor-
mation is the most important aspect. A common solution for all these three
situations is to cover a targeted area with a sensor’s footprint so as to gather
information about them. The technical term for solution to these classes of prob-
lems is called area coverage [1].
2. 2 Poulastya Mukherjee, Dr. Marek Schikora, Prof.Dr. Paul.G.Ploeger
Area coverage is usually done by attaching a sensor(e.g. camera, laser scanner)
to a robot and moving the robot in the target area in such a way, that informa-
tion about the entire area is captured by the robot’s sensor. Like most robotic
solutions, ground robots have been traditionally used for area coverage as well
because:
– they are easier to control
– they are robust
– they can carry higher loads compared to aerial vehicles which means they
can carry better battery,greater array of sensors, better computers etc.
However certain disadvantages of using ground robots in area coverage scenario
are:
– they are terrain dependent (although all terrain robots exists, they are slow
and might fail if the terrain is extremely uneven)
– they might not be able to overcome/bypass all obstacles
Herein lies the key advantage of Unmanned Aerial Vehicles or UAVs. Due to the
fact that UAVs have an extra degree of motion, they can avoid obstacles much
easier compared to ground robots. For disaster mitigation scenarios and defense
reconnaissance missions, only gathering information is not sufficient, gathering
information in the shortest possible time is also a key requirement. Being air-
borne, enables the UAVs to take the shortest distance to a goal - which might
not be always possible for a ground robot - thus saving a lot of time.
Recent advancements in electronics have led to significant increase in per-
formance of computers while both their size and weight have drastically fallen.
This has led to the development of UAVs with better on-board computers and
sensors, which have increased their scope of usage especially in area coverage
problems. However despite the technological advancements, UAVs still
– have lower computational ability
– and poor batteries
compared to even basic ground robots. Poor batteries directly translate to the
fact that UAVs cannot fly for a long time at a stretch. While limited computa-
tional abilities restricts it from running advanced algorithms (e.g. search algo-
rithms) on its on-board computer. The combination of these two factors pose a
serious challenge to UAVs being used for area coverage problems, which require a
robot to move over a large area for long duration of time while performing tasks
like recording data and avoiding obstacles. Thus developing efficient techniques
for doing area coverage using UAVs are a need of the hour.
A possible solution to the problem mentioned above as proposed in [1] is, using
multiple UAVs to do area coverage. It has the following advantages,
– reduction in the time required for each UAV to cover an area as now each
UAV will have to cover a smaller area
3. Cooperative Trajectory Planning Of UAVs For Area Coverage 3
– faster completion of task.
In this work we propose alternate algorithms, which generate a smooth mo-
tion trajectory like spirals, Lissajous Curves, hybrid approach. We also use two
UAVs for implementing our proposed methods, which will ensure reduced time
requirement to achieve goal while not making the process of controlling the over-
all multi-UAV, system too complex.
The aim of this work is to propose alternate methods - spirals, Lissajous
Curves, a hybrid approach- to the lawnmower approach [1] and randomized
approaches [2] and experimentally determine whether the proposed methods
are better or not. For implementing our proposed methods we will use Robot
Operating System(ROS) [3]. For simulating and visualizing our results we will
use Gazebo [4]
2 Problem formulation
Recent works in the field of area coverage using UAVs, is focused on two key
aspects for improving area coverage algorithms
– efficient methods of area decomposition [5–7]
– application of existing search based algorithms to generate coverage paths
[8–10, 2]
A key aspect missing from the above mentioned works is that these works do
not focus on the actual motion path followed by the UAVs, which in most cases
is simple back and forth motion or a lawnmower type approach, as proposed in
[1]. Though the proposed methods in [5–10, 2] ensure complete coverage, they
do not specify any means to reduce the time of coverage. Reducing the time of
area coverage is extremely important, to ensure the UAV is able to complete
the area coverage task within the time constraints imposed by its limited power
supply. Another problem in lawnmower approach is that, during the course of
the motion a UAV has to accelerate and decelerate multiple times ?? which
increases the time of coverage and also leads to higher power consumption.
Like any standard scientific work, we will carry out a state of the art eval-
uation of area coverage, focusing on area coverage using UAVs in section: ??.
Apart from area coverage using UAVs, we will also focus on multi UAV cooper-
ative systems and deployed libraries and existing platforms.
We will then describe our approach in section: 3 where we will focus on formu-
lating mathematical models for proposed algorithms. The following methods are
proposed
– Spiral method
– Lissajous Curves based method
– Hybrid approach
4. 4 Poulastya Mukherjee, Dr. Marek Schikora, Prof.Dr. Paul.G.Ploeger
For our implementation we will write a ROS package in C++ language which
will contain implementation of the proposed algorithms along with some other
key functionality for incorporating multi-agent behavior among the two UAVs.
We will also implement the lawnmower approach and Rapidly Exploring Ran-
dom Tree(RRT) approach. In the experiment section we will run each of these
algorithms while varying various parameters like size and shape of the area to
be covered, number of commanded waypoints. We will then formulate a cost
function to determine which method performs the best.
3 Geometry based trajectory planning
In this section we will briefly describe the approach taken to propose a solution
model for the problem formulation. Then we go on to propose an experimental
setup which will validate the proposed method.
3.1 Overview
A scientific approach to any problem, requires an extensive research on the state
of the art that exists for this particular problem, which has already been done
in the previous section. The first step is to determine on what factors does the
performance of the back and forth or lawnmower method -suggested in [1, 5, 11,
9, 2]- depend on, and record those parameters:
– Time of coverage
– Trace of the UAV’s path
– Velocity - both linear and angular
The next step is to determine algorithms which will minimize these parameters.
From [7] we learned that using simple trajectories like hemispherical - and cylin-
drical we can do coverage of any area. After implementing the algorithms which
will be formulated in 3.2, the next step will be to record the above mentioned
parameters and compare with the lawnmower approach parameters.
3.2 Formulating mathematical model of the algorithms
In [7] the authors propose hemispherical and cylindrical trajectories for 3D cov-
erage, likewise for 2D coverage we can use smooth curves like spirals, splines and
Lissajous Curves to generate trajectory path for the UAVs. In this section we
will describe the mathematical models of the above mentioned curves.
Spiral Spiral - According to the The American Heritage Dictionary of the En-
glish1
a spiral is defined as a curve which winds around a fixed center point at
1
The American Heritage Dictionary of the English Language, Houghton Mifflin Com-
pany, Fourth Edition, 2009.
5. Cooperative Trajectory Planning Of UAVs For Area Coverage 5
continuously increasing or decreasing radius. For generating way-points we need
to use the parametric equation of a spiral, given by2
x(t) = atcos(t) (1)
y(t) = atsin(t) (2)
In 1, 2 x and y are the coordinates of the waypoints, a is a constant value
and t is the varying radius. The value of a will depend on the dimensions of
the maximum value of x and y. The parameter t can be adjusted to increase or
decrease the width between two successive curves of the spiral. This value will
be adjusted according to the field of view radius of the UAV sensor.
Fig. 1. Archimedean Spiral 2
Lissajous Curves Lissajous Curves - From the work in [12] we get the para-
metric for of Lissajous curves.
x(t) = Acos(ωxt − δx) (3)
y(t) = Bcos(ωyt − δy) (4)
1
http://www.mathematische-basteleien.de/spiral.htm
6. 6 Poulastya Mukherjee, Dr. Marek Schikora, Prof.Dr. Paul.G.Ploeger
x(t) = Asin(ω ∗ t + δ) (5)
y(t) = Bsin(ω ∗ t) (6)
In equations 3, 4, 5, 6 x and y are the coordinates of the waypoints, ωxt and
ωyt are the amplitude of the cosine curve, ω and ω are the amplitudes of the
sine curve and δ, δx ,δy are the necessary phase difference to create a Lissajous
pattern and t is a variable. If the ratio of ω
ω = 5/4, the following Lissajous
pattern can be generated which can act as a space filling curve, thus in our case
act as an algorithm for coverage.
Fig. 2. Lissajous Curve with ω
ω
= 5 : 4 [12]
Hybrid of Spiral and Lissajous Curves To generate this Lissajous Curve
3, the ratio of ω
ω was set to 1 : 2, while varying the value of t(in this case
increasing) with each iteration we create the following coverage pattern. This
pattern can be used in case of doing coverage while persistently tracking any
target object. The center of the curve where the spirals intersect can be fixed on
a target object and the UAV can cover an area while persistently tracking the
target object.
7. Cooperative Trajectory Planning Of UAVs For Area Coverage 7
Fig. 3. Hybrid approach based on Spiral and Lissajous Curves method [12]
Rapidly Exploring Random Trees Along with the above mentioned geo-
metrical approaches we also implement a randomized approach in order to have
a better comparison of our proposed method. Unlike probabilistic approaches
which require point to point convergence to progress [13], we require an ap-
proach which can randomly expand and fill a coverage space. Rapidly Exploring
Random Trees are thus ideal for our problem[13]. The only change we make here
is that we do not consider a shortest path between start and goal node, instead
we try to generate a large number of nodes in the space to be covered and use
these generated nodes as waypoints for the UAVs to follow. We try to order the
waypoints generated in ascending order of Euclidean distance from the starting
point.
3.3 Formulating Criteria for Evaluation of Algorithms
In [1], the author defines three criteria for evaluating a coverage algorithm,
– percentage of area covered
– time taken by the UAVs to finish coverage of a given area
– computational and time complexity of the algorithm.
Along with these we add three more criteria:
– energy cost during its motion - which depends on the rate of change of
velocity
– percentage of area overlapped by sensor’s footprint.
– percentage of area exceeded from the commanded area
Percentage of Area covered In [1, 7] percentage of area covered is defined as
the actual area covered by the footprint of the robot’s sensor to the total goal
area that was to be covered. For calculating this, we record the pose of the UAV
8. 8 Poulastya Mukherjee, Dr. Marek Schikora, Prof.Dr. Paul.G.Ploeger
during its run. Next we get the value of field of view of the robot’s sensor from
the code. The field of view radius is calculated using the height of the robot and
the aperture angle of the sensor’s field of view. In the figure: 4 we illustrate an
example of a robot’s traced path and the corresponding trace path plus field of
view radius for the UAV.
Fig. 4. Two UAVs doing coverage of an area of 140x140 m2
using lawnmower approach
Time taken by the UAVs to complete coverage of a given area We
see that using either traditional search algorithms like A* [10] and randomized
approach like RRT algorithm [14],time complexity of the algorithms are low.
However these do not take into consideration the physical constraints of an
UAV. For example lets take the case of RRT algorithm, since the approach is
based on random generation of points, two consecutive points may be farther
from each other while the point after that may be near the initial point. This
will cause the UAV to move randomly and will end up having a high execution
time.
Computational and time complexity of the algorithm From the equations
of splines, Lissajous curves, hybrid curves, described in the approach section we
see that these methods can be implemented using a simple for loop which has a
time complexity of O(n) [15], where n is the maximum number of points to be
generated.
9. Cooperative Trajectory Planning Of UAVs For Area Coverage 9
Energy costs of UAV’s motion Since we are doing 2D coverage, the energy
cost of the UAV flying from ground to the required height is same for all the
methods mentioned above thus for calculating the energy costs we will be con-
sidering only Kinetic Energy.
From Newton’s 2nd law of motion we know that
F = m.a (7)
F = m.
dv
dt
(8)
We also know that work done W is given by
W = ∆KE....eq(7) (9)
where KE is Kinetic Energy. Therefore to minimize the change in Kinetic Energy
it is sufficient to show that the work done is minimum. Now,
W = F .x (10)
replacing from earlier equation of force
W = m.
dv
dt
.x (11)
considering m (mass) and x to be constant for a UAV and for a fixed area
covered, the only variable that can affect work done is the dv
dt factor.
Percentage of area overlap by sensor’s footprint A certain percentage
of area overlap might be a desired criteria in order to verify data gathered by
the UAV’s sensor, however unless it is specifically requested by the user, it is
undesirable as it causes a repetition of data by making an UAV fly over the same
region already covered by another UAV. Thus this causes an increase in total
energy cost.
Percentage of area exceeded as regards to the commanded area During
the course of its flight, if an UAV goes out of the commanded area then it is
considered as a negative trait. Though if the area exceeded is small, its effect
on the total energy cost wouldn’t be significant, however if the exceeded area is
large then it can seriously increase the energy cost.
3.4 Formulating the Performance Metric Function
Based on the parameters established in 4.1, we will define the performance
metric function.
10. 10 Poulastya Mukherjee, Dr. Marek Schikora, Prof.Dr. Paul.G.Ploeger
Assigning weights to the parameters Based on the importance of each
parameter we assign the following weights:
– Percentage of area coverage (w1) - This is a very important criteria for
measuring performance of an area coverage problem, hence we assign it a
value of w1 = 2
– Time taken by the UAV to complete coverage of a given area (w2) - This is
also a very important criteria for measuring performance of an area coverage
problem. We assign it a value of w2 = 2 as it is desired to have low time of
completion.
– Energy cost of UAV during coverage (w3) - Considering the fact that UAVs
have limited battery supply, keeping this parameter low is of utmost impor-
tance, hence we also assign it a value of w3 = 2.
– Percentage of area overlap (w4) - In some cases overlap might be a desired
criteria while in some cases not. Since this criteria does not play a big role
in affecting area coverage we assign it a value of w4 = 0.1 considering the
case we want to keep it as low as possible.
– Percentage of area exceeding commanded area of coverage (w5) - Exceeding
area of coverage is an undesirable factor and we want to keep it low. However
considering the overall performance, this factor is not as important as either
of percentage of area coverage, time to completion or energy cost, hence we
assign a value of w5 = 0.5
Formulating the final performance metric function We have seen in that
for the lawnmower approach the percentage of area coverage is 100%, percentage
of area overlap is 0% and percentage of area exceeded is also 0%, so we take
the values of the lawnmower approach as the standard for evaluating the other
approaches.
– For percentage of area coverage (denoted by P), 100% coverage is assigned
a value of 1. 90% coverage is assigned a value of 0.9 and likewise.
– We normalize time taken (denoted by t) to complete coverage for all ap-
proaches using the time for lawnmower approach.
– Similarly we normalize the energy cost (denoted by EC) of completing cov-
erage for all approaches using the energy cost for lawnmower approach.
– For percentage overlap (denoted by O) we consider no overlap to have a
value of 0 and 100% overlap to have a value of 1.
– Similarly for percentage of area exceeded (denoted by E) we consider no
area exceeded to have a value of 0 and 100% exceeded area to have a value
of 1
The final function is defined as
fp(method) = w1 ∗ (Pmethod) + w2 ∗ (1 −
tmethod
tlawnmower
) (12)
+w3 ∗ (1 −
ECmethod
EClawnmower
) + w4 ∗ (1 − Omethod) + w5 ∗ (1 − Emethod) (13)
11. Cooperative Trajectory Planning Of UAVs For Area Coverage 11
4 Experimentation and Results
4.1 Conducting Experiments
We will conduct our experiments in a simulated environment. The simulation
platform is Gazebo while the code is written using Robot Operating System(ROS)
and C++. Using the ROS package Hector Quadrotor we spawn two UAVs in-
side the rolling landscape world of Gazebo. Then we run our coverage quadrotor
nodes (one node for each UAV) and record run time, poses, velocity information
from the ground state topic. For each algorithm - Lawnmower, Spiral, Lissajous
curves, Hybrid approach, RRT - we vary the the area to be covered.
4.2 Results of the Experiments
In this section we will analyze the results of the experiment which were presented
in the previous section.
Percentage of area coverage and parameters which affect it From the
images presented , we saw that there is a relationship between the commanded
area to be covered and the percentage of area actually covered by the UAV’s
sensor. We present a table which lists percentage of area coverage to the size of
the total area to be covered.
Table 1. Area vs Percentage Area Coverage
Method 80x80 100x100 160x120 140x140 140x340
Lawnmower 100% 100% 100% 100% 100%
Spiral 93% 96% 97% 98% 99%
Lissajous Curves 100% 99% 96% 94% 92%
Hybrid 99% 99% 99% 99% 99%
RRT 40% 37% 31% 28% 25%
From the results in table: 1 we see that the proposed algorithms - Spiral,
Hybrid achieves almost complete coverage and their performance remains mostly
unchanged with the size of the area to be covered, while the performance of
Lissajous curves method degrades with larger areas. The main reason for the
degradation of the performance of the Lissajous curves method is due to the
non-uniform gap between the generated paths.
Another conclusion that can be drawn from the above data is that,the per-
formance of the spiral method improves as the size of the area to be covered
increases, even though the improvement is quite nominal.
12. 12 Poulastya Mukherjee, Dr. Marek Schikora, Prof.Dr. Paul.G.Ploeger
Fig. 5. Incomplete coverage of the Lissajous curves method due to non-uniform gap
between generated paths for area of 140x140 m2
Completion time of the proposed methods In this section we will analyze
the time performance of the algorithms compared to the size of the area to be
covered.
Table 2. Table of Area of coverage in m2
to Time taken in seconds for all methods
Method 80x80 m2
100x100 m2
160x120 m2
140x140 m2
140x340 m2
Lawnmower 437 secs 440 secs 864 secs 918 secs 1703 secs
Spiral 420 secs 428 secs 674 secs 706 secs 1043 secs
Lissajous Curves 420 secs 426 secs 672 secs 705 secs 1042 secs
Hybrid 421 secs 427 secs 675 secs 706 secs 1041 secs
RRT 372 secs 368 secs 568 secs 602 secs 997 secs
From the data in the table 2 it is quite evident that as the size of the
area to be covered increases, the time to required to cover it using either of
Spiral, Lissajous curves, or Hybrid method is significantly lower than that of
Lawnmower method. Even though the proposed method may cover slightly lesser
area compared to the lawnmower method, the significant reduction in time makes
these methods more efficient compared to the Lawnmower approach.
13. Cooperative Trajectory Planning Of UAVs For Area Coverage 13
Energy Consumed calculation for each method In this section we will
calculate the energy consumed for areas of (i)80x80 m2
, (iii) 160x120 m2
and
(iii) 140x140 m2
We assume the weight of an UAV to be m = 10 kg.
Using the formula in section: 3.3, we calculate the energy costs. The results
are presented in table 3
Table 3. Table for energy cost for each method.
Method 80x80 m2
160x120 m2
140x140 m2
Lawnmower 291.01 kJ 1155.71 kJ 1280.52 kJ
Spiral 227.65 kJ 577.12 kJ 639.91 kJ
Lissajous Curves 917.65 kJ 1464.11 kJ 1869.93 kJ
Hybrid 603.93 kJ 951.43 kJ 1122.45 kJ
RRT 158.11 kJ 363.55 kJ 423.55 kJ
From the data in table 3 we see that Spiral method has the least energy
cost and as the size of the area to be covered increases, its energy consumption
compared to other methods become even lower. Though the least energy con-
sumption is by the RRT method, it does not do coverage of the area completely
and thus we are not considering it for our final evaluation.
4.3 Calculating the Performance Metric
In section: 4.2 the results of all the parameters were presented. Now using those
parameters we will calculate the values of the performance metric function for
each method for (i)80x80 m2
,(ii)160x120 m2
and (iii) 140x140 m2
.
Table 4. Performance Metric of methods based on area
Method 80x80 m2
160x120 m2
140x140 m2
Lawnmower 2.6 2.6 2.6
Spiral 2.9 3.94 3.98
Lissajous Curves 0.26 2.24 1.83
Hybrid 0.45 3.23 3.32
RRT 0.01 0.78 1.45
From table 4, it is quite evident that the spiral method performs best com-
pared to all other methods and its performance increases as the size of the area
to be covered increases. The next best method is the hybrid method, whose value
also increases as the size of the area to be covered increases, but it is also to be
noted that for smaller areas, its performance is very poor.
14. 14 Poulastya Mukherjee, Dr. Marek Schikora, Prof.Dr. Paul.G.Ploeger
5 Conclusion and Future Work
In this work we proposed a geometry based approach to area coverage problem
using UAVs. We proposed that since UAVs have limited computational capa-
bilities, using algorithms which require very little computational power can be
helpful in creating better solutions. Based on the experiments carried out we
came to the conclusion that for doing coverage of small areas lawnmower ap-
proach is more suited as it ensures complete coverage in minimal time, however
from the results in the previous section, we can clearly conclude that spiral
method is the best method overall as it ensures atleast 98 % coverage while hav-
ing minimum energy cost and time to completion. Along with the spiral method,
the hybrid method also shows some promise, but its performance in small areas
is really poor. The RRT method does not guarantee complete coverage.
Regarding the fact that a smooth motion of the UAVs during its course of flight
is a requirement not only for capturing information uniformly but also ensuring
a high level of flight stability, the results published in clearly show that the spiral
method gives the best performance as it ensures its acceleration and deceleration
are within a certain range thus ensuring greater flight stability
A very important future work is to implement the proposed algorithms on a
real system and determine whether their performance is in accordance with the
findings of this work. In terms of improving the performance of the algorithm,
introducing machine learning to make the UAVs autonomously decide which
algorithm to use for which scenario will be a big improvement. The primary
requirement for creating such a system would be to run multiple number of test
to gather enough data.
Apart from the above mentioned directions, changing the orientation and po-
sition of the sensor on the UAV and studying how it affect coverage performance
is also a possible extension of this work.
References
1. Howie Choset. Coverage For Robotics - A Survey Of Recent Results. Annals of
Mathematics and Artificial Intelligence, 31:113 – 126, 2001.
2. Wei-Long Yang, Luo Lei, and Jing-Sheng Deng. Optimization and Improvement for
Multi-UAV Cooperative Reconnaissance Mission Planning Problem. In Interna-
tional Computer Conference on Wavelet Active Media Technology and Information
Processing (ICCWAMTIP), 2014 11th, pages 10–15, Chengdu, China, 2014. IEEE.
3. Morgan Quigley, Ken Conley, Brian Gerkey, Josh Faust, Tully Foote, Jeremy Leibs,
Rob Wheeler, and Andrew Y Ng. ROS: an open-source Robot Operating System.
In ICRA workshop on open source software, volume 3, page 5, Kobe, Japan, 2009.
4. Nathan Koenig and Andrew Howard. Design and Use Paradigms for Gazebo, an
Open-Source Multi-robot Simulator. In International Conference on Intelligent
Robots and Systems, 2004.(IROS 2004). Proceedings. 2004 IEEE/RSJ, volume 3,
pages 2149–2154, Sendai, Japan, 2004. IEEE.
15. Cooperative Trajectory Planning Of UAVs For Area Coverage 15
5. JF Araujo, PB Sujit, and JB Sousa. Multiple UAV Area Decomposition and
Coverage. In 2013 IEEE Symposium on Computational Intelligence for Security
and Defense Applications (CISDA), pages 30–37, Singapore, 2013. IEEE.
6. Alessio Del Bue, M Tamassia, F Signorini, Vittorio Murino, and A Farinelli. Visual
Coverage Using Autonomous Mobile Robots for Search and Rescue Applications.
In 2013 IEEE International Symposium on Safety, Security, and Rescue Robotics
(SSRR), pages 1–8, Linkping, Sweden, 2013. IEEE.
7. Peng Cheng, James Keller, and Vijay Kumar. Time-optimal UAV Trajectory Plan-
ning for 3D Urban Structure Coverage. In IEEE/RSJ International Conference on
Intelligent Robots and Systems, 2008. IROS 2008., pages 2750–2757, Nice, France,
2008. IEEE.
8. Yong Bao, Xiaowei Fu, and Xiaoguang Gao. Path Planning for Reconnaissance
UAV Based on Particle Swarm Optimization. In Second International Conference
on Computational Intelligence and Natural Computing Proceedings (CINC), 2010,
volume 2, pages 28–32, Wuhan, China, 2010. IEEE.
9. Hyo-Sang Shin, C´edric Leboucher, and Antonios Tsourdos. Resource Allocation
with Cooperative Path Planning for Multiple UAVs. In UKACC International Con-
ference on Control (CONTROL), 2012, pages 298–303, Coventry University,UK,
2012. IEEE.
10. Wu Xinggang, Guo Cong, and Li Yibo. Variable Probability Based Bidirectional
RRT algorithm for UAV Path Planning. In Control and Decision Conference (2014
CCDC), The 26th Chinese, pages 2217–2222, China, 2014. IEEE.
11. Jie Chen, Wenzhong Zha, Zhihong Peng, and Jian Zhang. Cooperative Area Re-
connaissance for Multi-UAV in Dynamic Environment. In Control Conference
(ASCC), 2013 9th Asian, pages 1–6, Istanbul, 2013. IEEE.
12. H Cundy and A Rollett. Lissajouss Figures. Mathematical Models, pages 242–244,
1989.
13. Steven M. LaValle. Rapidly Exploring Random Trees A New Tool for Path Plan-
ning. In Proceedings of the Mediterranean Conference on Control and Automation,
Sardinia, Italy, 1998.
14. Zhu W Li L Teng L Chenglong Y Jiaxun K. Enhanced Sparse A * Search for UAV
Path Planning Using Dubins Path Estimation. In Proceedings of the 33rd Chinese
Control Conference. China, 2014.
15. Michael Sipser. Introduction to the Theory of Computation. Thomson Course
Technology, Boston, 2006.