SlideShare a Scribd company logo
T O C S S
         B B  O B


                  Hannes Schulz

               University of Freiburg, ACS



                      Feb 2008
O


  1   I: C S  R

  2   T M I  U G S
        Idea 1: Use Config Space Structure
        Idea 2: Increase Connectivity


  3   S A

  4   E: C  S S

  5   C  C
O


  1   I: C S  R

  2   T M I  U G S
        Idea 1: Use Config Space Structure
        Idea 2: Increase Connectivity


  3   S A

  4   E: C  S S

  5   C  C
Intro        Two Main Ideas         Sampling Algorithm   Experiments    Comments and Conclusion


 C S

        World Space                                             Configuration Space




                              http://ford.ieor.berkeley.edu/cspace
Intro        Two Main Ideas         Sampling Algorithm   Experiments    Comments and Conclusion


 C S

        World Space                                             Configuration Space




                              http://ford.ieor.berkeley.edu/cspace
Intro        Two Main Ideas         Sampling Algorithm   Experiments    Comments and Conclusion


 C S

        World Space                                             Configuration Space




                              http://ford.ieor.berkeley.edu/cspace
Intro        Two Main Ideas         Sampling Algorithm   Experiments    Comments and Conclusion


 C S

        World Space                                             Configuration Space




                              http://ford.ieor.berkeley.edu/cspace
Intro        Two Main Ideas         Sampling Algorithm   Experiments    Comments and Conclusion


 C S

        World Space                                             Configuration Space




                              http://ford.ieor.berkeley.edu/cspace
Intro        Two Main Ideas         Sampling Algorithm   Experiments    Comments and Conclusion


 C S

        World Space                                             Configuration Space




                              http://ford.ieor.berkeley.edu/cspace
Intro        Two Main Ideas         Sampling Algorithm   Experiments    Comments and Conclusion


 C S

        World Space                                             Configuration Space




                              http://ford.ieor.berkeley.edu/cspace
Intro        Two Main Ideas         Sampling Algorithm   Experiments    Comments and Conclusion


 C S

        World Space                                             Configuration Space




                              http://ford.ieor.berkeley.edu/cspace
Intro        Two Main Ideas         Sampling Algorithm   Experiments    Comments and Conclusion


 C S

        World Space                                             Configuration Space




                              http://ford.ieor.berkeley.edu/cspace
Intro     Two Main Ideas   Sampling Algorithm   Experiments   Comments and Conclusion


 R M




        Config Space
        w/ Obstacles
Intro     Two Main Ideas   Sampling Algorithm   Experiments   Comments and Conclusion


 R M




        Config Space
        w/ Obstacles
        and Samples
Intro     Two Main Ideas   Sampling Algorithm   Experiments   Comments and Conclusion


 R M




        Config Space                Visibility Road Map
        w/ Obstacles
        and Samples
Intro     Two Main Ideas   Sampling Algorithm   Experiments   Comments and Conclusion


 R M




        Config Space                Visibility Road Map
        w/ Obstacles                  Planned Path
        and Samples
Intro     Two Main Ideas   Sampling Algorithm   Experiments   Comments and Conclusion


 R M




        Config Space                Visibility Road Map
        w/ Obstacles                  Planned Path
        and Samples

         How to sample quickly in high-dimensional Config Space?
Intro   Two Main Ideas   Sampling Algorithm   Experiments   Comments and Conclusion


 S A


        S

                     G
            Uniform
Intro   Two Main Ideas   Sampling Algorithm       Experiments   Comments and Conclusion


 S A


        S                           S

                     G                        G
            Uniform         Wavefront
                               single query
Intro   Two Main Ideas   Sampling Algorithm       Experiments        Comments and Conclusion


 S A


        S                           S

                     G                        G
            Uniform         Wavefront                Model- Guided
                               single query                 multi-query
Intro   Two Main Ideas   Sampling Algorithm       Experiments        Comments and Conclusion


 S A


        S                           S

                     G                        G
            Uniform         Wavefront                Model- Guided
                               single query                 multi-query
Intro   Two Main Ideas   Sampling Algorithm       Experiments        Comments and Conclusion


 S A


        S                           S                    Entropy-guided,
                                                          Model-guided,
                                                         Bridge-Sampling,
                     G                        G                 ...


            Uniform         Wavefront                           Guided
                               single query                 multi-query
Intro   Two Main Ideas   Sampling Algorithm       Experiments            Comments and Conclusion


 S A


        S                           S                    Entropy-guided,
                                                          Model-guided,
                                                         Bridge-Sampling,
                     G                        G                 ...


            Uniform         Wavefront                           Guided
                               single query                 multi-query

                                                      In this paper:


                                                                   ?
                                                    “utility-guided”
                                                                multi-query
O


  1   I: C S  R

  2   T M I  U G S
        Idea 1: Use Config Space Structure
        Idea 2: Increase Connectivity


  3   S A

  4   E: C  S S

  5   C  C
O


  1   I: C S  R

  2   T M I  U G S
        Idea 1: Use Config Space Structure
        Idea 2: Increase Connectivity


  3   S A

  4   E: C  S S

  5   C  C
Intro   Two Main Ideas   Sampling Algorithm   Experiments   Comments and Conclusion


 U C S S F A L
Intro     Two Main Ideas   Sampling Algorithm   Experiments   Comments and Conclusion


 U C S S F A L


        Obstacle Sample



        Free Space Sample
Intro   Two Main Ideas   Sampling Algorithm   Experiments   Comments and Conclusion


 U C S S F A L
Intro    Two Main Ideas   Sampling Algorithm   Experiments   Comments and Conclusion


 U C S S F A L




        Function Approximator: Approximate Model of Config Space
        Use Model to select next free sample
        Using all known samples aids active learning
O


  1   I: C S  R

  2   T M I  U G S
        Idea 1: Use Config Space Structure
        Idea 2: Increase Connectivity


  3   S A

  4   E: C  S S

  5   C  C
Intro         Two Main Ideas   Sampling Algorithm    Experiments   Comments and Conclusion


 D  “U”   S

               Component 1
                                          Obstacle

                                                    Component 2




        Entropy: Probability that random sample is in visibility region of
        particular component
Intro        Two Main Ideas   Sampling Algorithm    Experiments   Comments and Conclusion


 D  “U”   S

              Component 1
                                         Obstacle

                                                   Component 2




        Red Sample: Entropy unchanged, Zero information gain
Intro         Two Main Ideas   Sampling Algorithm    Experiments   Comments and Conclusion


 D  “U”   S

               Component 1
                                          Obstacle

                                                    Component 2




        Entropy: Probability that random sample is in visibility region of
        particular component
Intro         Two Main Ideas   Sampling Algorithm   Experiments   Comments and Conclusion


 D  “U”   S


                                          Obstacle




                                    Just 1 Component left
        Red Sample: Less Entropy, Large information gain, high Utility
O


  1   I: C S  R

  2   T M I  U G S
        Idea 1: Use Config Space Structure
        Idea 2: Increase Connectivity


  3   S A

  4   E: C  S S

  5   C  C
Intro         Two Main Ideas   Sampling Algorithm     Experiments   Comments and Conclusion


 E C U M  C

        Component 1
                                    Obstacle

                                                    Component 2




        Application Idea 2: Try to increase connectivity
Intro         Two Main Ideas   Sampling Algorithm     Experiments   Comments and Conclusion


 E C U M  C

        Component 1
                                    Obstacle

                                                    Component 2




        Application Idea 2: Try to increase connectivity
Intro     Two Main Ideas   Sampling Algorithm     Experiments   Comments and Conclusion


 E C U M  C

        Component 1
                                Obstacle

                                                Component 2

                                 Center Point
Intro     Two Main Ideas   Sampling Algorithm     Experiments   Comments and Conclusion


 E C U M  C

        Component 1
                                Obstacle

                                                Component 2
Intro         Two Main Ideas   Sampling Algorithm     Experiments   Comments and Conclusion


 W  P S  M

        Component 1
                                    Obstacle

                                                    Component 2




        Application Idea 1: Exploit model of config space
Intro         Two Main Ideas   Sampling Algorithm     Experiments   Comments and Conclusion


 W  P S  M

        Component 1
                                    Obstacle

                                                    Component 2




        Application Idea 1: Exploit model of config space
Intro     Two Main Ideas   Sampling Algorithm     Experiments   Comments and Conclusion


 W  P S  M

        Component 1
                                Obstacle

                                                Component 2
O


  1   I: C S  R

  2   T M I  U G S
        Idea 1: Use Config Space Structure
        Idea 2: Increase Connectivity


  3   S A

  4   E: C  S S

  5   C  C
Intro      Two Main Ideas   Sampling Algorithm    Experiments    Comments and Conclusion


 R




        3 or 4 Joints with 3 DOF                 Mobile base (2 DOF)
        each                                     2 Joints with 1 / 2 DOF each
        9 DOF / 12 DOF                           4 DOF / 6 DOF
          Compare only Sampling strategy until path found
          Difficulty: Analyzing Overhead of Model, Utility Evaluation
Intro        Two Main Ideas   Sampling Algorithm   Experiments   Comments and Conclusion


 R

        Runtimes: 4-DOF mobile manipulator
Intro        Two Main Ideas   Sampling Algorithm   Experiments   Comments and Conclusion


 R

        Fraction of Config Space covered: 9-DOF arm
O


  1   I: C S  R

  2   T M I  U G S
        Idea 1: Use Config Space Structure
        Idea 2: Increase Connectivity


  3   S A

  4   E: C  S S

  5   C  C
Intro         Two Main Ideas   Sampling Algorithm   Experiments       Comments and Conclusion


 A P P  U-G S?




                    Goal                                          Start




        Initially, just two Samples at start and goal, respectively. What
        happens?
Intro        Two Main Ideas   Sampling Algorithm   Experiments       Comments and Conclusion


 A P P  U-G S?




                   Goal                                          Start




        Line between two clusters
Intro        Two Main Ideas   Sampling Algorithm   Experiments       Comments and Conclusion


 A P P  U-G S?




                   Goal                                          Start




        Sample candidates around mitpoint
Intro        Two Main Ideas   Sampling Algorithm   Experiments       Comments and Conclusion


 A P P  U-G S?




                   Goal                                          Start




        Model does not provide information, choose any.
Intro        Two Main Ideas   Sampling Algorithm   Experiments       Comments and Conclusion


 A P P  U-G S?




                   Goal                                          Start




        Green: New model. Suppose same nodes chosen again. What
        happens?
Intro        Two Main Ideas   Sampling Algorithm   Experiments       Comments and Conclusion


 A P P  U-G S?




                   Goal                                          Start




        New points cluster around previous points
Intro        Two Main Ideas     Sampling Algorithm   Experiments       Comments and Conclusion


 A P P  U-G S?




                   Goal                                            Start




        “Worst case” scenario
Intro        Two Main Ideas     Sampling Algorithm   Experiments   Comments and Conclusion


 C



        Burns & Brock introduced a sampling algorithm

            for multi-query planning
            uses active learning
            maximizes utility
            outperforms other algorithms
            not thoroughly evaluated
            may have strong dependency on parameters/environment
Intro   Two Main Ideas   Sampling Algorithm   Experiments   Comments and Conclusion


 D




                                        ?

More Related Content

Viewers also liked

HH6 How the construction of large-scale dams is changing the hydropolitical r...
HH6 How the construction of large-scale dams is changing the hydropolitical r...HH6 How the construction of large-scale dams is changing the hydropolitical r...
HH6 How the construction of large-scale dams is changing the hydropolitical r...
Ana Cascao
 
Dades i operadors
Dades i operadorsDades i operadors
Dades i operadors
Alex Muntada Duran
 
Invoicing
InvoicingInvoicing
Invoicing
TeachStreet
 
Callme
CallmeCallme
Callme
sutrisno2629
 
December 2009 TeachStreet Teacher Webinar & Meet-up
December 2009 TeachStreet Teacher Webinar & Meet-upDecember 2009 TeachStreet Teacher Webinar & Meet-up
December 2009 TeachStreet Teacher Webinar & Meet-up
TeachStreet
 
Ad Campaign For Calmax
Ad Campaign For CalmaxAd Campaign For Calmax
Ad Campaign For Calmax
Karan Chengappa
 
Search: The Purest Form of Interaction Design
Search: The Purest Form of Interaction DesignSearch: The Purest Form of Interaction Design
Search: The Purest Form of Interaction Design
Chiara Fox Ogan
 
Jaladas De Awebowey
Jaladas De AweboweyJaladas De Awebowey
Jaladas De Awebowey
Adrián Chávez
 
Cascao_South_Sudan_Nile_Hydropolitics
Cascao_South_Sudan_Nile_HydropoliticsCascao_South_Sudan_Nile_Hydropolitics
Cascao_South_Sudan_Nile_Hydropolitics
Ana Cascao
 
Entrepreneur! london 2012 slideshare
Entrepreneur! london 2012 slideshareEntrepreneur! london 2012 slideshare
Entrepreneur! london 2012 slideshare
Erik Scarcia
 
How to craft consolidated math notes
How to craft consolidated math notesHow to craft consolidated math notes
How to craft consolidated math notes
colwilliamson
 
A life without complaints
A life without complaintsA life without complaints
A life without complaints
sutrisno2629
 
Cascao&Conway Stockholm Doube Faced Cooperation Nile Basin
Cascao&Conway Stockholm Doube Faced Cooperation Nile BasinCascao&Conway Stockholm Doube Faced Cooperation Nile Basin
Cascao&Conway Stockholm Doube Faced Cooperation Nile Basin
Ana Cascao
 
Vladi funtzioak
Vladi funtzioakVladi funtzioak
Vladi funtzioak
Josu Jimenez Maia
 
Awebowey!!! Agosto2009
Awebowey!!! Agosto2009Awebowey!!! Agosto2009
Awebowey!!! Agosto2009
Adrián Chávez
 
Jack London Background
Jack London BackgroundJack London Background
Jack London Background
prhahn
 
Iui protocol -draft
Iui protocol -draftIui protocol -draft
Iui protocol -draft
Avinash Bhatt
 

Viewers also liked (18)

HH6 How the construction of large-scale dams is changing the hydropolitical r...
HH6 How the construction of large-scale dams is changing the hydropolitical r...HH6 How the construction of large-scale dams is changing the hydropolitical r...
HH6 How the construction of large-scale dams is changing the hydropolitical r...
 
Dades i operadors
Dades i operadorsDades i operadors
Dades i operadors
 
Invoicing
InvoicingInvoicing
Invoicing
 
Callme
CallmeCallme
Callme
 
December 2009 TeachStreet Teacher Webinar & Meet-up
December 2009 TeachStreet Teacher Webinar & Meet-upDecember 2009 TeachStreet Teacher Webinar & Meet-up
December 2009 TeachStreet Teacher Webinar & Meet-up
 
Ad Campaign For Calmax
Ad Campaign For CalmaxAd Campaign For Calmax
Ad Campaign For Calmax
 
Search: The Purest Form of Interaction Design
Search: The Purest Form of Interaction DesignSearch: The Purest Form of Interaction Design
Search: The Purest Form of Interaction Design
 
Jaladas De Awebowey
Jaladas De AweboweyJaladas De Awebowey
Jaladas De Awebowey
 
Cascao_South_Sudan_Nile_Hydropolitics
Cascao_South_Sudan_Nile_HydropoliticsCascao_South_Sudan_Nile_Hydropolitics
Cascao_South_Sudan_Nile_Hydropolitics
 
Entrepreneur! london 2012 slideshare
Entrepreneur! london 2012 slideshareEntrepreneur! london 2012 slideshare
Entrepreneur! london 2012 slideshare
 
How to craft consolidated math notes
How to craft consolidated math notesHow to craft consolidated math notes
How to craft consolidated math notes
 
A life without complaints
A life without complaintsA life without complaints
A life without complaints
 
Cascao&Conway Stockholm Doube Faced Cooperation Nile Basin
Cascao&Conway Stockholm Doube Faced Cooperation Nile BasinCascao&Conway Stockholm Doube Faced Cooperation Nile Basin
Cascao&Conway Stockholm Doube Faced Cooperation Nile Basin
 
030413
030413030413
030413
 
Vladi funtzioak
Vladi funtzioakVladi funtzioak
Vladi funtzioak
 
Awebowey!!! Agosto2009
Awebowey!!! Agosto2009Awebowey!!! Agosto2009
Awebowey!!! Agosto2009
 
Jack London Background
Jack London BackgroundJack London Background
Jack London Background
 
Iui protocol -draft
Iui protocol -draftIui protocol -draft
Iui protocol -draft
 

Similar to Toward Optimal Configuration Space Sampling

Learning Kinematics from direct Self-Observation using Nearest-Neighbour Methods
Learning Kinematics from direct Self-Observation using Nearest-Neighbour MethodsLearning Kinematics from direct Self-Observation using Nearest-Neighbour Methods
Learning Kinematics from direct Self-Observation using Nearest-Neighbour Methods
cijat
 
Andrew rusling 21 experiments to increase velocity
Andrew rusling 21 experiments to increase velocityAndrew rusling 21 experiments to increase velocity
Andrew rusling 21 experiments to increase velocity
Scrum Australia Pty Ltd
 
21 experiments to increase velocity
21 experiments to increase velocity21 experiments to increase velocity
21 experiments to increase velocity
Andrew Rusling
 
Managing Uncertainty in Value-based SE
Managing Uncertainty in Value-based SEManaging Uncertainty in Value-based SE
Managing Uncertainty in Value-based SE
CS, NcState
 
Arbonne's Results Presentation
Arbonne's Results PresentationArbonne's Results Presentation
Arbonne's Results Presentation
guest06b488
 
The deep bootstrap framework review
The deep bootstrap framework reviewThe deep bootstrap framework review
The deep bootstrap framework review
taeseon ryu
 
Test Driven Development
Test Driven DevelopmentTest Driven Development
Test Driven Development
guy_davis
 
Web 2.0 架站工具—AJAX By Examples-馮彥文(Tempo)
Web 2.0 架站工具—AJAX By Examples-馮彥文(Tempo)Web 2.0 架站工具—AJAX By Examples-馮彥文(Tempo)
Web 2.0 架站工具—AJAX By Examples-馮彥文(Tempo)
taiwanweb20
 
2009 05 01 How To Build A Lean Startup Step By Step
2009 05 01 How To Build A Lean Startup Step By Step2009 05 01 How To Build A Lean Startup Step By Step
2009 05 01 How To Build A Lean Startup Step By Step
Eric Ries
 
21 Experiments to Increase Throughput
21 Experiments to Increase Throughput21 Experiments to Increase Throughput
21 Experiments to Increase Throughput
Andrew Rusling
 
Computational methods applications in air pollution modeling (Dr. Yadghar)
Computational methods applications in air pollution modeling (Dr. Yadghar)Computational methods applications in air pollution modeling (Dr. Yadghar)
Computational methods applications in air pollution modeling (Dr. Yadghar)
knowdiff
 

Similar to Toward Optimal Configuration Space Sampling (11)

Learning Kinematics from direct Self-Observation using Nearest-Neighbour Methods
Learning Kinematics from direct Self-Observation using Nearest-Neighbour MethodsLearning Kinematics from direct Self-Observation using Nearest-Neighbour Methods
Learning Kinematics from direct Self-Observation using Nearest-Neighbour Methods
 
Andrew rusling 21 experiments to increase velocity
Andrew rusling 21 experiments to increase velocityAndrew rusling 21 experiments to increase velocity
Andrew rusling 21 experiments to increase velocity
 
21 experiments to increase velocity
21 experiments to increase velocity21 experiments to increase velocity
21 experiments to increase velocity
 
Managing Uncertainty in Value-based SE
Managing Uncertainty in Value-based SEManaging Uncertainty in Value-based SE
Managing Uncertainty in Value-based SE
 
Arbonne's Results Presentation
Arbonne's Results PresentationArbonne's Results Presentation
Arbonne's Results Presentation
 
The deep bootstrap framework review
The deep bootstrap framework reviewThe deep bootstrap framework review
The deep bootstrap framework review
 
Test Driven Development
Test Driven DevelopmentTest Driven Development
Test Driven Development
 
Web 2.0 架站工具—AJAX By Examples-馮彥文(Tempo)
Web 2.0 架站工具—AJAX By Examples-馮彥文(Tempo)Web 2.0 架站工具—AJAX By Examples-馮彥文(Tempo)
Web 2.0 架站工具—AJAX By Examples-馮彥文(Tempo)
 
2009 05 01 How To Build A Lean Startup Step By Step
2009 05 01 How To Build A Lean Startup Step By Step2009 05 01 How To Build A Lean Startup Step By Step
2009 05 01 How To Build A Lean Startup Step By Step
 
21 Experiments to Increase Throughput
21 Experiments to Increase Throughput21 Experiments to Increase Throughput
21 Experiments to Increase Throughput
 
Computational methods applications in air pollution modeling (Dr. Yadghar)
Computational methods applications in air pollution modeling (Dr. Yadghar)Computational methods applications in air pollution modeling (Dr. Yadghar)
Computational methods applications in air pollution modeling (Dr. Yadghar)
 

More from cijat

CRF-Filters: Discriminative Particle Filters for Sequential State Estimation
CRF-Filters: Discriminative Particle Filters for Sequential State EstimationCRF-Filters: Discriminative Particle Filters for Sequential State Estimation
CRF-Filters: Discriminative Particle Filters for Sequential State Estimation
cijat
 
The ball is not just orange.
The ball is not just orange.The ball is not just orange.
The ball is not just orange.
cijat
 
RoboCup Introduction
RoboCup IntroductionRoboCup Introduction
RoboCup Introduction
cijat
 
Local and Global Gating of Synaptic Plasticity
Local and Global Gating of Synaptic PlasticityLocal and Global Gating of Synaptic Plasticity
Local and Global Gating of Synaptic Plasticity
cijat
 
Auxiliary
AuxiliaryAuxiliary
Auxiliary
cijat
 
Affordances
AffordancesAffordances
Affordances
cijat
 

More from cijat (6)

CRF-Filters: Discriminative Particle Filters for Sequential State Estimation
CRF-Filters: Discriminative Particle Filters for Sequential State EstimationCRF-Filters: Discriminative Particle Filters for Sequential State Estimation
CRF-Filters: Discriminative Particle Filters for Sequential State Estimation
 
The ball is not just orange.
The ball is not just orange.The ball is not just orange.
The ball is not just orange.
 
RoboCup Introduction
RoboCup IntroductionRoboCup Introduction
RoboCup Introduction
 
Local and Global Gating of Synaptic Plasticity
Local and Global Gating of Synaptic PlasticityLocal and Global Gating of Synaptic Plasticity
Local and Global Gating of Synaptic Plasticity
 
Auxiliary
AuxiliaryAuxiliary
Auxiliary
 
Affordances
AffordancesAffordances
Affordances
 

Recently uploaded

Programming Foundation Models with DSPy - Meetup Slides
Programming Foundation Models with DSPy - Meetup SlidesProgramming Foundation Models with DSPy - Meetup Slides
Programming Foundation Models with DSPy - Meetup Slides
Zilliz
 
Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)
Jakub Marek
 
JavaLand 2024: Application Development Green Masterplan
JavaLand 2024: Application Development Green MasterplanJavaLand 2024: Application Development Green Masterplan
JavaLand 2024: Application Development Green Masterplan
Miro Wengner
 
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
Jeffrey Haguewood
 
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
Edge AI and Vision Alliance
 
Driving Business Innovation: Latest Generative AI Advancements & Success Story
Driving Business Innovation: Latest Generative AI Advancements & Success StoryDriving Business Innovation: Latest Generative AI Advancements & Success Story
Driving Business Innovation: Latest Generative AI Advancements & Success Story
Safe Software
 
5th LF Energy Power Grid Model Meet-up Slides
5th LF Energy Power Grid Model Meet-up Slides5th LF Energy Power Grid Model Meet-up Slides
5th LF Energy Power Grid Model Meet-up Slides
DanBrown980551
 
HCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAU
HCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAUHCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAU
HCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAU
panagenda
 
Deep Dive: AI-Powered Marketing to Get More Leads and Customers with HyperGro...
Deep Dive: AI-Powered Marketing to Get More Leads and Customers with HyperGro...Deep Dive: AI-Powered Marketing to Get More Leads and Customers with HyperGro...
Deep Dive: AI-Powered Marketing to Get More Leads and Customers with HyperGro...
saastr
 
Fueling AI with Great Data with Airbyte Webinar
Fueling AI with Great Data with Airbyte WebinarFueling AI with Great Data with Airbyte Webinar
Fueling AI with Great Data with Airbyte Webinar
Zilliz
 
Freshworks Rethinks NoSQL for Rapid Scaling & Cost-Efficiency
Freshworks Rethinks NoSQL for Rapid Scaling & Cost-EfficiencyFreshworks Rethinks NoSQL for Rapid Scaling & Cost-Efficiency
Freshworks Rethinks NoSQL for Rapid Scaling & Cost-Efficiency
ScyllaDB
 
Best 20 SEO Techniques To Improve Website Visibility In SERP
Best 20 SEO Techniques To Improve Website Visibility In SERPBest 20 SEO Techniques To Improve Website Visibility In SERP
Best 20 SEO Techniques To Improve Website Visibility In SERP
Pixlogix Infotech
 
TrustArc Webinar - 2024 Global Privacy Survey
TrustArc Webinar - 2024 Global Privacy SurveyTrustArc Webinar - 2024 Global Privacy Survey
TrustArc Webinar - 2024 Global Privacy Survey
TrustArc
 
Skybuffer SAM4U tool for SAP license adoption
Skybuffer SAM4U tool for SAP license adoptionSkybuffer SAM4U tool for SAP license adoption
Skybuffer SAM4U tool for SAP license adoption
Tatiana Kojar
 
Public CyberSecurity Awareness Presentation 2024.pptx
Public CyberSecurity Awareness Presentation 2024.pptxPublic CyberSecurity Awareness Presentation 2024.pptx
Public CyberSecurity Awareness Presentation 2024.pptx
marufrahmanstratejm
 
WeTestAthens: Postman's AI & Automation Techniques
WeTestAthens: Postman's AI & Automation TechniquesWeTestAthens: Postman's AI & Automation Techniques
WeTestAthens: Postman's AI & Automation Techniques
Postman
 
Generating privacy-protected synthetic data using Secludy and Milvus
Generating privacy-protected synthetic data using Secludy and MilvusGenerating privacy-protected synthetic data using Secludy and Milvus
Generating privacy-protected synthetic data using Secludy and Milvus
Zilliz
 
Introduction of Cybersecurity with OSS at Code Europe 2024
Introduction of Cybersecurity with OSS  at Code Europe 2024Introduction of Cybersecurity with OSS  at Code Europe 2024
Introduction of Cybersecurity with OSS at Code Europe 2024
Hiroshi SHIBATA
 
HCL Notes and Domino License Cost Reduction in the World of DLAU
HCL Notes and Domino License Cost Reduction in the World of DLAUHCL Notes and Domino License Cost Reduction in the World of DLAU
HCL Notes and Domino License Cost Reduction in the World of DLAU
panagenda
 
Trusted Execution Environment for Decentralized Process Mining
Trusted Execution Environment for Decentralized Process MiningTrusted Execution Environment for Decentralized Process Mining
Trusted Execution Environment for Decentralized Process Mining
LucaBarbaro3
 

Recently uploaded (20)

Programming Foundation Models with DSPy - Meetup Slides
Programming Foundation Models with DSPy - Meetup SlidesProgramming Foundation Models with DSPy - Meetup Slides
Programming Foundation Models with DSPy - Meetup Slides
 
Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)
 
JavaLand 2024: Application Development Green Masterplan
JavaLand 2024: Application Development Green MasterplanJavaLand 2024: Application Development Green Masterplan
JavaLand 2024: Application Development Green Masterplan
 
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
 
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
 
Driving Business Innovation: Latest Generative AI Advancements & Success Story
Driving Business Innovation: Latest Generative AI Advancements & Success StoryDriving Business Innovation: Latest Generative AI Advancements & Success Story
Driving Business Innovation: Latest Generative AI Advancements & Success Story
 
5th LF Energy Power Grid Model Meet-up Slides
5th LF Energy Power Grid Model Meet-up Slides5th LF Energy Power Grid Model Meet-up Slides
5th LF Energy Power Grid Model Meet-up Slides
 
HCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAU
HCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAUHCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAU
HCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAU
 
Deep Dive: AI-Powered Marketing to Get More Leads and Customers with HyperGro...
Deep Dive: AI-Powered Marketing to Get More Leads and Customers with HyperGro...Deep Dive: AI-Powered Marketing to Get More Leads and Customers with HyperGro...
Deep Dive: AI-Powered Marketing to Get More Leads and Customers with HyperGro...
 
Fueling AI with Great Data with Airbyte Webinar
Fueling AI with Great Data with Airbyte WebinarFueling AI with Great Data with Airbyte Webinar
Fueling AI with Great Data with Airbyte Webinar
 
Freshworks Rethinks NoSQL for Rapid Scaling & Cost-Efficiency
Freshworks Rethinks NoSQL for Rapid Scaling & Cost-EfficiencyFreshworks Rethinks NoSQL for Rapid Scaling & Cost-Efficiency
Freshworks Rethinks NoSQL for Rapid Scaling & Cost-Efficiency
 
Best 20 SEO Techniques To Improve Website Visibility In SERP
Best 20 SEO Techniques To Improve Website Visibility In SERPBest 20 SEO Techniques To Improve Website Visibility In SERP
Best 20 SEO Techniques To Improve Website Visibility In SERP
 
TrustArc Webinar - 2024 Global Privacy Survey
TrustArc Webinar - 2024 Global Privacy SurveyTrustArc Webinar - 2024 Global Privacy Survey
TrustArc Webinar - 2024 Global Privacy Survey
 
Skybuffer SAM4U tool for SAP license adoption
Skybuffer SAM4U tool for SAP license adoptionSkybuffer SAM4U tool for SAP license adoption
Skybuffer SAM4U tool for SAP license adoption
 
Public CyberSecurity Awareness Presentation 2024.pptx
Public CyberSecurity Awareness Presentation 2024.pptxPublic CyberSecurity Awareness Presentation 2024.pptx
Public CyberSecurity Awareness Presentation 2024.pptx
 
WeTestAthens: Postman's AI & Automation Techniques
WeTestAthens: Postman's AI & Automation TechniquesWeTestAthens: Postman's AI & Automation Techniques
WeTestAthens: Postman's AI & Automation Techniques
 
Generating privacy-protected synthetic data using Secludy and Milvus
Generating privacy-protected synthetic data using Secludy and MilvusGenerating privacy-protected synthetic data using Secludy and Milvus
Generating privacy-protected synthetic data using Secludy and Milvus
 
Introduction of Cybersecurity with OSS at Code Europe 2024
Introduction of Cybersecurity with OSS  at Code Europe 2024Introduction of Cybersecurity with OSS  at Code Europe 2024
Introduction of Cybersecurity with OSS at Code Europe 2024
 
HCL Notes and Domino License Cost Reduction in the World of DLAU
HCL Notes and Domino License Cost Reduction in the World of DLAUHCL Notes and Domino License Cost Reduction in the World of DLAU
HCL Notes and Domino License Cost Reduction in the World of DLAU
 
Trusted Execution Environment for Decentralized Process Mining
Trusted Execution Environment for Decentralized Process MiningTrusted Execution Environment for Decentralized Process Mining
Trusted Execution Environment for Decentralized Process Mining
 

Toward Optimal Configuration Space Sampling

  • 1. T O C S S B B  O B Hannes Schulz University of Freiburg, ACS Feb 2008
  • 2. O 1 I: C S  R 2 T M I  U G S Idea 1: Use Config Space Structure Idea 2: Increase Connectivity 3 S A 4 E: C  S S 5 C  C
  • 3. O 1 I: C S  R 2 T M I  U G S Idea 1: Use Config Space Structure Idea 2: Increase Connectivity 3 S A 4 E: C  S S 5 C  C
  • 4. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion C S World Space Configuration Space http://ford.ieor.berkeley.edu/cspace
  • 5. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion C S World Space Configuration Space http://ford.ieor.berkeley.edu/cspace
  • 6. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion C S World Space Configuration Space http://ford.ieor.berkeley.edu/cspace
  • 7. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion C S World Space Configuration Space http://ford.ieor.berkeley.edu/cspace
  • 8. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion C S World Space Configuration Space http://ford.ieor.berkeley.edu/cspace
  • 9. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion C S World Space Configuration Space http://ford.ieor.berkeley.edu/cspace
  • 10. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion C S World Space Configuration Space http://ford.ieor.berkeley.edu/cspace
  • 11. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion C S World Space Configuration Space http://ford.ieor.berkeley.edu/cspace
  • 12. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion C S World Space Configuration Space http://ford.ieor.berkeley.edu/cspace
  • 13. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion R M Config Space w/ Obstacles
  • 14. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion R M Config Space w/ Obstacles and Samples
  • 15. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion R M Config Space Visibility Road Map w/ Obstacles and Samples
  • 16. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion R M Config Space Visibility Road Map w/ Obstacles Planned Path and Samples
  • 17. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion R M Config Space Visibility Road Map w/ Obstacles Planned Path and Samples How to sample quickly in high-dimensional Config Space?
  • 18. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion S A S G Uniform
  • 19. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion S A S S G G Uniform Wavefront single query
  • 20. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion S A S S G G Uniform Wavefront Model- Guided single query multi-query
  • 21. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion S A S S G G Uniform Wavefront Model- Guided single query multi-query
  • 22. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion S A S S Entropy-guided, Model-guided, Bridge-Sampling, G G ... Uniform Wavefront Guided single query multi-query
  • 23. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion S A S S Entropy-guided, Model-guided, Bridge-Sampling, G G ... Uniform Wavefront Guided single query multi-query In this paper: ? “utility-guided” multi-query
  • 24. O 1 I: C S  R 2 T M I  U G S Idea 1: Use Config Space Structure Idea 2: Increase Connectivity 3 S A 4 E: C  S S 5 C  C
  • 25. O 1 I: C S  R 2 T M I  U G S Idea 1: Use Config Space Structure Idea 2: Increase Connectivity 3 S A 4 E: C  S S 5 C  C
  • 26. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion U C S S F A L
  • 27. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion U C S S F A L Obstacle Sample Free Space Sample
  • 28. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion U C S S F A L
  • 29. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion U C S S F A L Function Approximator: Approximate Model of Config Space Use Model to select next free sample Using all known samples aids active learning
  • 30. O 1 I: C S  R 2 T M I  U G S Idea 1: Use Config Space Structure Idea 2: Increase Connectivity 3 S A 4 E: C  S S 5 C  C
  • 31. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion D  “U”   S Component 1 Obstacle Component 2 Entropy: Probability that random sample is in visibility region of particular component
  • 32. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion D  “U”   S Component 1 Obstacle Component 2 Red Sample: Entropy unchanged, Zero information gain
  • 33. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion D  “U”   S Component 1 Obstacle Component 2 Entropy: Probability that random sample is in visibility region of particular component
  • 34. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion D  “U”   S Obstacle Just 1 Component left Red Sample: Less Entropy, Large information gain, high Utility
  • 35. O 1 I: C S  R 2 T M I  U G S Idea 1: Use Config Space Structure Idea 2: Increase Connectivity 3 S A 4 E: C  S S 5 C  C
  • 36. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion E C U M  C Component 1 Obstacle Component 2 Application Idea 2: Try to increase connectivity
  • 37. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion E C U M  C Component 1 Obstacle Component 2 Application Idea 2: Try to increase connectivity
  • 38. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion E C U M  C Component 1 Obstacle Component 2 Center Point
  • 39. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion E C U M  C Component 1 Obstacle Component 2
  • 40. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion W  P S  M Component 1 Obstacle Component 2 Application Idea 1: Exploit model of config space
  • 41. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion W  P S  M Component 1 Obstacle Component 2 Application Idea 1: Exploit model of config space
  • 42. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion W  P S  M Component 1 Obstacle Component 2
  • 43. O 1 I: C S  R 2 T M I  U G S Idea 1: Use Config Space Structure Idea 2: Increase Connectivity 3 S A 4 E: C  S S 5 C  C
  • 44. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion R 3 or 4 Joints with 3 DOF Mobile base (2 DOF) each 2 Joints with 1 / 2 DOF each 9 DOF / 12 DOF 4 DOF / 6 DOF Compare only Sampling strategy until path found Difficulty: Analyzing Overhead of Model, Utility Evaluation
  • 45. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion R Runtimes: 4-DOF mobile manipulator
  • 46. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion R Fraction of Config Space covered: 9-DOF arm
  • 47. O 1 I: C S  R 2 T M I  U G S Idea 1: Use Config Space Structure Idea 2: Increase Connectivity 3 S A 4 E: C  S S 5 C  C
  • 48. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion A P P  U-G S? Goal Start Initially, just two Samples at start and goal, respectively. What happens?
  • 49. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion A P P  U-G S? Goal Start Line between two clusters
  • 50. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion A P P  U-G S? Goal Start Sample candidates around mitpoint
  • 51. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion A P P  U-G S? Goal Start Model does not provide information, choose any.
  • 52. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion A P P  U-G S? Goal Start Green: New model. Suppose same nodes chosen again. What happens?
  • 53. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion A P P  U-G S? Goal Start New points cluster around previous points
  • 54. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion A P P  U-G S? Goal Start “Worst case” scenario
  • 55. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion C Burns & Brock introduced a sampling algorithm for multi-query planning uses active learning maximizes utility outperforms other algorithms not thoroughly evaluated may have strong dependency on parameters/environment
  • 56. Intro Two Main Ideas Sampling Algorithm Experiments Comments and Conclusion D ?