Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
AUTONOMOUS SYSTEMS LABORATORY | ICT CENTRE
Dr. Raja Jurdak
Towards Continental Scale Tracking of
Flying Foxes
Continental Scale Tracking
• Track the position and state of small assets for
long durations
Continental Scale Flying Fox ...
Continental Scale Tracking
• Track the position and state of small assets for
long durations
• Why is it important for Aus...
Continental Scale Tracking
• Track the position and state of small assets for
long durations
• Why is it important for Aus...
Continental Scale Tracking
The fundamental ICT Challenge
• Need to use energy hungry GPS
• Operate within very tight energ...
National Flying Fox Monitoring Program
Funding of ~ $5M by Federal, state
governments and CSIRO – 3 years
Why track them
•...
National Flying Fox Monitoring Program
Funding of ~ $5M by Federal, state
governments and CSIRO – 3 years
Why track them
•...
Continental Scale Flying Fox Monitoring| Raja Jurdak8 |
Current trackers work well.. but not for long
A day in the life of...
Continental-scale tracking
Goals
• Near perpetual tracking across Australia
• Discovery of new roosting camps
Continental ...
Goals
• Near perpetual tracking across Australia
• Discovery of new roosting camps
Continental Scale Flying Fox Monitoring...
Goals
• Near perpetual tracking across Australia
• Discovery of new roosting camps
The fundamental challenge
Long-term loc...
Camazotz
Continental Scale Flying Fox Monitoring| Raja Jurdak12 |
• Multimodal
sensing platform
• Low power SoC
R. Jurdak,...
Communication Dependencies on 3D Mobility
Continental Scale Flying Fox Monitoring| Raja Jurdak13 |
Communication Dependencies - Antenna
Continental Scale Flying Fox Monitoring| Raja Jurdak14 |
Communication Dependencies - Speed
Continental Scale Flying Fox Monitoring| Raja Jurdak15 |
Communication Dependencies - Angle
Continental Scale Flying Fox Monitoring| Raja Jurdak16 |
Altitude using relative air pressure
Continental Scale Flying Fox Monitoring| Raja Jurdak17 |
GPS vertical accuracy is +/-...
Altitude using relative air pressure
Continental Scale Flying Fox Monitoring| Raja Jurdak18 |
GPS vertical accuracy is +/-...
Characterizing GPS Performance
19 | Continental Scale Flying Fox Monitoring| Raja Jurdak
Characterizing GPS Performance
20 |
• R. Jurdak, P. Corke, A. Cotillon, et al.,
"Energy-efficient Localisation: GPS Duty
C...
Characterizing GPS Performance
Continental Scale Flying Fox Monitoring| Raja Jurdak21 |
• R. Jurdak, P. Corke, A. Cotillon...
Characterizing GPS Performance
Continental Scale Flying Fox Monitoring| Raja Jurdak22 |
Energy Profiling
Continental Scale Flying Fox Monitoring| Raja Jurdak23 |
Energy Profiling
Continental Scale Flying Fox Monitoring| Raja Jurdak24 |
GPS samples are a
precious resources (can
take >...
Sensor-triggered GPS Sampling
Continental Scale Flying Fox Monitoring| Raja Jurdak25 |
• Use one or more of the cheap on-b...
Understanding activities
• Videos
Continental Scale Flying Fox Monitoring| Raja Jurdak26 |
Sensor-triggered GPS samples (Accelerometer)
• Compute average
vector at rest  gravity
• Compute angle
between current ve...
Sensor-triggered GPS samples (Audio)
Continental Scale Flying Fox Monitoring| Raja Jurdak28 |
• Frequency peaks at 2-4Khz
...
Sensor-triggered GPS samples (Audio)
Continental Scale Flying Fox Monitoring| Raja Jurdak29 |
• Frequency peaks at 2-4Khz
...
Multimodal Event dissociation
• When one sensor is
insufficient to capture
event-of-interest
• Example: how to
dissociate ...
Multimodal Event dissociation
• When one sensor is
insufficient to capture
event-of-interest
• Example: how to
dissociate ...
Multimodal Activity-based Localisation
Collaredevents Nearbyevents Powerconsumption
DetectedEvents
AveragePowerConsumption...
Where to from here?
Continental Scale Flying Fox Monitoring| Raja Jurdak33 |
Open Challenges
• Delay-tolerant …
• Data storage (what to store or not)
• Sampling (maximum information for energy buck)
...
Mobility Modeling
• Establish mobility
dependencies
• Temporal
• Spatial
• Environmental
• Social
• Use to drive GPS sampl...
Cooperative GPS Sampling
• Determine co-location among multiple foxes
• Cooperative GPS sampling in a group
Continental Sc...
Sample-and-Process
• Sample GPS pseudorange for 1 ms *
• Use nearby landmark to estimate location
Continental Scale Flying...
Conclusion
• Next steps
• Deeper investigation into GPS dynamics
• Modeling mobility
• Progressively longer field trials –...
AUTONOMOUS SYSTEMS LABORATORY | ICT CENTRE
Dr. Raja Jurdak
Research Group Leader, Pervasive Computing
Principal Research S...
Upcoming SlideShare
Loading in …5
×

Towards Continental-scale Tracking of Flying Foxes

138 views

Published on

Long-term outdoor localisation with battery-powered devices remains an unsolved challenge, mainly due to the high energy consumption of GPS modules. The use of inertial sensors and short-range radio can reduce reliance on GPS to prolong the operational lifetime of tracking devices, but they only provide coarse-grained control over GPS activity. An alternative yet promising approach is to use context-sensitive mobility models to guide scheduling and sampling decisions in localisation algorithms. In this talk, I will present our work towards continental-scale long-term tracking of flying foxes, as part of the National Flying Fox Monitoring Program, using a model-driven approach. At the core of our approach is the multimodal GPS-enabled Camazotz sensor node platform that has been designed at CSIRO for flying fox collars, with a cumulative weight of below 30g. The talk will cover our recent experience with trialling these platforms in the field on live flying foxes to collect multimodal sensor data for developing models of their mobility. I will also discuss the road ahead for designing adaptive model-driven algorithms for energy-efficient localisation.

Published in: Science
  • Be the first to comment

Towards Continental-scale Tracking of Flying Foxes

  1. 1. AUTONOMOUS SYSTEMS LABORATORY | ICT CENTRE Dr. Raja Jurdak Towards Continental Scale Tracking of Flying Foxes
  2. 2. Continental Scale Tracking • Track the position and state of small assets for long durations Continental Scale Flying Fox Monitoring| Raja Jurdak2 |
  3. 3. Continental Scale Tracking • Track the position and state of small assets for long durations • Why is it important for Australia? • Sparse population • Large landmass • Agriculture and biosecurity Continental Scale Flying Fox Monitoring| Raja Jurdak3 |
  4. 4. Continental Scale Tracking • Track the position and state of small assets for long durations • Why is it important for Australia? • Sparse population • Large landmass • Agriculture and biosecurity • Relevant applications • Asset tracking • Livestock tracking • Wildlife tracking Continental Scale Flying Fox Monitoring| Raja Jurdak4 |
  5. 5. Continental Scale Tracking The fundamental ICT Challenge • Need to use energy hungry GPS • Operate within very tight energy budgets <Project Title> | <Project Lead>5 | Continental Scale Flying Fox Monitoring| Raja Jurdak
  6. 6. National Flying Fox Monitoring Program Funding of ~ $5M by Federal, state governments and CSIRO – 3 years Why track them • Disease vectors – Hendra cost $20M/year – Ebola in Asia/Africa – Coronavirus (KSA - 2013) • Seed dispersal agents • Threatened species? • Not well understood Continental Scale Flying Fox Monitoring| Raja Jurdak
  7. 7. National Flying Fox Monitoring Program Funding of ~ $5M by Federal, state governments and CSIRO – 3 years Why track them • Disease vectors – Hendra cost $20M/year – Ebola in Asia/Africa – Coronavirus (KSA - 2013) • Seed dispersal agents • Threatened species? • Not well understood What to track • Habitat use • Individual interactions • FF/Animal interactions Continental Scale Flying Fox Monitoring| Raja Jurdak
  8. 8. Continental Scale Flying Fox Monitoring| Raja Jurdak8 | Current trackers work well.. but not for long A day in the life of a flying fox
  9. 9. Continental-scale tracking Goals • Near perpetual tracking across Australia • Discovery of new roosting camps Continental Scale Flying Fox Monitoring| Raja Jurdak
  10. 10. Goals • Near perpetual tracking across Australia • Discovery of new roosting camps Continental Scale Flying Fox Monitoring| Raja Jurdak Phase 1 Track for >6 months with 100m accuracy Phase 2 Track for >6 months with 10m accuracy Phase 3 Track for >6 months with 5m accuracy Continental-scale tracking
  11. 11. Goals • Near perpetual tracking across Australia • Discovery of new roosting camps The fundamental challenge Long-term localisation with tiny energy budget • Weight (30-50g) • Mobility (up to 100km/night) • Truly remote (continental scale) • Intermittent connectivity Continental Scale Flying Fox Monitoring| Raja Jurdak Phase 1 Track for >6 months with 100m accuracy Phase 2 Track for >6 months with 10m accuracy Phase 3 Track for >6 months with 5m accuracy Continental-scale tracking
  12. 12. Camazotz Continental Scale Flying Fox Monitoring| Raja Jurdak12 | • Multimodal sensing platform • Low power SoC R. Jurdak, P. Sommer, B. Kusy, N. Kottege, C. Crossman, A. McKeown, D. Westcott, “Multimodal Activity-based GPS Sampling," To appear in proceedings of the 12th International Conference on Information Processing in Sensor Networks (IPSN), Philadelphia, USA, April, 2013.
  13. 13. Communication Dependencies on 3D Mobility Continental Scale Flying Fox Monitoring| Raja Jurdak13 |
  14. 14. Communication Dependencies - Antenna Continental Scale Flying Fox Monitoring| Raja Jurdak14 |
  15. 15. Communication Dependencies - Speed Continental Scale Flying Fox Monitoring| Raja Jurdak15 |
  16. 16. Communication Dependencies - Angle Continental Scale Flying Fox Monitoring| Raja Jurdak16 |
  17. 17. Altitude using relative air pressure Continental Scale Flying Fox Monitoring| Raja Jurdak17 | GPS vertical accuracy is +/-22m (datasheets)
  18. 18. Altitude using relative air pressure Continental Scale Flying Fox Monitoring| Raja Jurdak18 | GPS vertical accuracy is +/-22m (datasheets)
  19. 19. Characterizing GPS Performance 19 | Continental Scale Flying Fox Monitoring| Raja Jurdak
  20. 20. Characterizing GPS Performance 20 | • R. Jurdak, P. Corke, A. Cotillon, et al., "Energy-efficient Localisation: GPS Duty Cycling with Radio Ranging," To appear in ACM TOSN: 9(2), May 2013. (in press) • R. Jurdak, P. Corke, D. Dharman, and G. Salagnac. "Adaptive GPS Duty Cycling and Radio Ranging for Energy-Efficient Localization," In proceedings of ACM Sensys, pp. 57-70. Zurich, Switzerland, November 2010. Continental Scale Flying Fox Monitoring| Raja Jurdak
  21. 21. Characterizing GPS Performance Continental Scale Flying Fox Monitoring| Raja Jurdak21 | • R. Jurdak, P. Corke, A. Cotillon, et al., "Energy-efficient Localisation: GPS Duty Cycling with Radio Ranging," To appear in ACM TOSN: 9(2), May 2013. (in press) • R. Jurdak, P. Corke, D. Dharman, and G. Salagnac. "Adaptive GPS Duty Cycling and Radio Ranging for Energy-Efficient Localization," In proceedings of ACM Sensys, pp. 57-70. Zurich, Switzerland, November 2010.
  22. 22. Characterizing GPS Performance Continental Scale Flying Fox Monitoring| Raja Jurdak22 |
  23. 23. Energy Profiling Continental Scale Flying Fox Monitoring| Raja Jurdak23 |
  24. 24. Energy Profiling Continental Scale Flying Fox Monitoring| Raja Jurdak24 | GPS samples are a precious resources (can take >30 seconds) How do we schedule the samples to capture movement patterns at minimum energy cost?
  25. 25. Sensor-triggered GPS Sampling Continental Scale Flying Fox Monitoring| Raja Jurdak25 | • Use one or more of the cheap on-board sensors to detect activities of interest and trigger GPS samples • Some activities of interest
  26. 26. Understanding activities • Videos Continental Scale Flying Fox Monitoring| Raja Jurdak26 |
  27. 27. Sensor-triggered GPS samples (Accelerometer) • Compute average vector at rest  gravity • Compute angle between current vector and gravity • Detect sustained angular shifts above 90o • 100% accuracy in detecting 11 true events • Video footage as ground truth Continental Scale Flying Fox Monitoring| Raja Jurdak27 | 1.4 1.5 1.6 1.7 1.8 1.9 2 x10 5 −2 0 2 4 Sample Accelerationprojectionon meanvector(G) 1.4 1.5 1.6 1.7 1.8 1.9 2 x10 5 0 100 200 Sample Angle−currentand gravity(degrees)
  28. 28. Sensor-triggered GPS samples (Audio) Continental Scale Flying Fox Monitoring| Raja Jurdak28 | • Frequency peaks at 2-4Khz • Lightweight features are based on calculating the mean signal energy and counting the number of zero crossings of a 1024 sample sliding window with an overlap of 50% • Video footage as ground truth
  29. 29. Sensor-triggered GPS samples (Audio) Continental Scale Flying Fox Monitoring| Raja Jurdak29 | • Frequency peaks at 2-4Khz • Lightweight features are based on calculating the mean signal energy and counting the number of zero crossings of a 1024 sample sliding window with an overlap of 50% • Video footage as ground truth
  30. 30. Multimodal Event dissociation • When one sensor is insufficient to capture event-of-interest • Example: how to dissociate interaction events involving a collared animal from interaction events involving nearby animals only? Continental Scale Flying Fox Monitoring| Raja Jurdak30 |
  31. 31. Multimodal Event dissociation • When one sensor is insufficient to capture event-of-interest • Example: how to dissociate interaction events involving a collared animal from interaction events involving nearby animals only? Continental Scale Flying Fox Monitoring| Raja Jurdak31 |
  32. 32. Multimodal Activity-based Localisation Collaredevents Nearbyevents Powerconsumption DetectedEvents AveragePowerConsumption(mW) Accelerometer MAL collared only MAL nearby only MAL all events 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 9 8 7 6 5 4 2 0 1 3 Audio Continental Scale Flying Fox Monitoring| Raja Jurdak32 | L ocali sat ion A p p r oach A n im al int er act ion s C ollar ed A ll D issociat ed Duty cycled GPS X A cceleromet er-t riggered X A udio-t riggered X A ccel. A ND A udio X A ccel. OR A udio X X Table 5: M A L can det ect all event s and dissociat e int er act ion event involving collared animal or near by animals. in our simulations. We compare a baseline approach of a duty cycled GPS with a period of 20s with triggered GPS sampling approaches based on the accelerometer only, audio only, or on the combination of audio and accelerometer sen- sors. We group all detected ground truth interactions into events that meet the 25s to 1min duration constraint. A successful detection in our simulation is when the algorithm obtains at least one GPS sample during the event. During the given time window, the duty cycled GPS mod- ule remains active for a total of 451s (including lock times) and successfully obtains GPS samples during each of the four events of interest, yielding an overall node power con- sumption of around 33mW. Figure 13 summarises the re- sults of sensor-triggered GPS sampling. The accelerometer- triggered GPS manages to detect only two events (only the events from the collared bat) with a cumulative GPS active Collaredevents Nearbyevents Powerconsum DetectedEvents Accelerometer MAL collared only MAL nearby only MAL all even 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 Audio Figur e 13: Per for mance of M A L acceler om et er - and audio-t rigger ed GP can be t uned t o capt ur e eit her int er act io of t he collar ed animal, or nearby int er act i only. M A L can also det ect and dissoc types of int er act ion event s wit h compar ab consumpt ion t o audio. alongside GPS. The ZebraNet project [5] reports position records for zebras every few minutes. I make the energy problem more tractable Zebra include a solar panel, which assume that the pan silient to normal animal activities. Positioning GPS only, and the nodes propagate their infor flooding in order to facilitate data acquisition by sink. Dyo e al. [3] use a heterogeneous sensor ne
  33. 33. Where to from here? Continental Scale Flying Fox Monitoring| Raja Jurdak33 |
  34. 34. Open Challenges • Delay-tolerant … • Data storage (what to store or not) • Sampling (maximum information for energy buck) • Communication (priorities, fairness, throughput) • Energy management (consumption, harvesting, prediction) • Tradeoffs? • Mobility model-driven sampling • How to build the model without the data  adaptive models • How flexible do these models need to be? Continental Scale Flying Fox Monitoring| Raja Jurdak34 |
  35. 35. Mobility Modeling • Establish mobility dependencies • Temporal • Spatial • Environmental • Social • Use to drive GPS sampling • Use percolation theory to explore overlaps between information and disease spread Continental Scale Flying Fox Monitoring| Raja Jurdak35 |
  36. 36. Cooperative GPS Sampling • Determine co-location among multiple foxes • Cooperative GPS sampling in a group Continental Scale Flying Fox Monitoring| Raja Jurdak36 | • R. Jurdak, B. Kusy, and A. Cotillon, "Group-based Motion Detection for Energy-efficient Localization," Journal of Sensor and Actuator Networks. 1(3):183-216, October 2012. (Invited paper) • R. Jurdak, P. Corke, A. Cotillon, et al., "Energy-efficient Localisation: GPS Duty Cycling with Radio Ranging," To appear in ACM TOSN: 9(2), May 2013. (in press) • R. Jurdak, P. Corke, D. Dharman, and G. Salagnac. "Adaptive GPS Duty Cycling and Radio Ranging for Energy-Efficient Localization," In proceedings of the ACM Sensys, pp. 57-70. Zurich, Switzerland, November 2010.
  37. 37. Sample-and-Process • Sample GPS pseudorange for 1 ms * • Use nearby landmark to estimate location Continental Scale Flying Fox Monitoring| Raja Jurdak37 | loops. So, once a GPS produces its first location fix, sub- sequent location estimates become fast. However, once the GPS receiver stops tracking, the utility of previously known Doppler shifts and code phases diminishes quickly. Typi- cally, after 30 seconds of non-tracking, the GPS receiver has to start all over again. Correlation Figure3. An example of acquisition result. Drawbacks • Postfacto locations only • Large Amounts of data/fix Opportunities • Explore design space online/offline • Explore compressive sensing to reduce data/fix * Jie Liu, Bodhi Priyantha, Ted Hart, Heitor Ramos, Antonio A.F. Loureiro, and Qiang Wang. Energy efficient gps sensing with cloud offloading. In Proc. SenSys, November 2012.
  38. 38. Conclusion • Next steps • Deeper investigation into GPS dynamics • Modeling mobility • Progressively longer field trials – 30, 150, 1000 nodes • Live monitoring • Continental Scale Tracking • Near-perpetual monitoring of position and condition • Very challenging yet interesting research problem with real application drivers Continental Scale Flying Fox Monitoring| Raja Jurdak38 |
  39. 39. AUTONOMOUS SYSTEMS LABORATORY | ICT CENTRE Dr. Raja Jurdak Research Group Leader, Pervasive Computing Principal Research Scientist rjurdak@ieee.org Thank You

×