Real-Time Stochastic Optimal Control for Multi-Agent Quadrotor Systems

Real-Time Stochastic Optimal Control for Multi-Agent Quadrotor Systems. Gómez, V., Thijssen, S., Symington, A., Hailes, S., & Kappen, B. In

Paper abstract bibtex

This paper presents a novel method for controlling teams of unmanned aerial vehicles using Stochastic Optimal Control (SOC) theory. The approach consists of a centralized high-level controller that computes optimal state trajectories as velocity sequences, and a platform-specific low-level controller which ensures that these velocity sequences are met. The high-level control task is expressed as a centralized path-integral control problem, for which optimal control computation corresponds to a probabilistic inference problem that can be solved by efficient sampling methods. Through simulation we show that our SOC approach (a) has significant benefits compared to deterministic control and other SOC methods in multi-modal problems with noise-dependent optimal solutions, (b) is capable of controlling a large number of platforms in real-time, and (c) yields collective emergent behavior in the form of flight formations. Finally, we show that our approach works for real platforms, by controlling a team of three quadrotors.

@inproceedings {icaps16-98,
    track    = {Robotics Track},
    title    = {Real-Time Stochastic Optimal Control for Multi-Agent Quadrotor Systems},
    url      = {http://www.aaai.org/ocs/index.php/ICAPS/ICAPS16/paper/view/13105},
    author   = {Vicenç Gómez and  Sep Thijssen and  Andrew Symington and  Stephen Hailes and  Bert Kappen},
    abstract = {This paper presents a novel method for controlling teams of unmanned aerial vehicles using Stochastic Optimal Control (SOC) theory. The approach consists of a centralized high-level controller that computes optimal state trajectories as velocity sequences, and a platform-specific low-level controller which ensures that these velocity sequences are met. The high-level control task is expressed as a centralized path-integral control problem, for which optimal control computation corresponds to a probabilistic inference problem that can be solved by efficient sampling methods. Through simulation we show that our SOC approach (a) has significant benefits compared to deterministic control and other SOC methods in multi-modal problems with noise-dependent optimal solutions, (b) is capable of controlling a large number of platforms in real-time, and (c) yields collective emergent behavior in the form of flight formations. Finally, we show that our approach works for real platforms, by controlling a team of three quadrotors.},
    keywords = {planning and coordination methods for multiple robots,robot motion; path; task and mission planning and execution,acquisition of planning models for robotics,integrated planning and execution in robotic architectures,real-world planning applications for autonomous robots}
}

Downloads: 0

{"_id":"bYjmi6HCTXPfHuFLg","bibbaseid":"gmez-thijssen-symington-hailes-kappen-realtimestochasticoptimalcontrolformultiagentquadrotorsystems","downloads":1,"creationDate":"2016-03-09T03:08:50.231Z","title":"Real-Time Stochastic Optimal Control for Multi-Agent Quadrotor Systems","author_short":["Gómez, V.","Thijssen, S.","Symington, A.","Hailes, S.","Kappen, B."],"year":null,"bibtype":"inproceedings","biburl":"icaps16.icaps-conference.org/papers.bib","bibdata":{"bibtype":"inproceedings","type":"inproceedings","track":"Robotics Track","title":"Real-Time Stochastic Optimal Control for Multi-Agent Quadrotor Systems","url":"http://www.aaai.org/ocs/index.php/ICAPS/ICAPS16/paper/view/13105","author":[{"firstnames":["Vicenç"],"propositions":[],"lastnames":["Gómez"],"suffixes":[]},{"firstnames":["Sep"],"propositions":[],"lastnames":["Thijssen"],"suffixes":[]},{"firstnames":["Andrew"],"propositions":[],"lastnames":["Symington"],"suffixes":[]},{"firstnames":["Stephen"],"propositions":[],"lastnames":["Hailes"],"suffixes":[]},{"firstnames":["Bert"],"propositions":[],"lastnames":["Kappen"],"suffixes":[]}],"abstract":"This paper presents a novel method for controlling teams of unmanned aerial vehicles using Stochastic Optimal Control (SOC) theory. The approach consists of a centralized high-level controller that computes optimal state trajectories as velocity sequences, and a platform-specific low-level controller which ensures that these velocity sequences are met. The high-level control task is expressed as a centralized path-integral control problem, for which optimal control computation corresponds to a probabilistic inference problem that can be solved by efficient sampling methods. Through simulation we show that our SOC approach (a) has significant benefits compared to deterministic control and other SOC methods in multi-modal problems with noise-dependent optimal solutions, (b) is capable of controlling a large number of platforms in real-time, and (c) yields collective emergent behavior in the form of flight formations. Finally, we show that our approach works for real platforms, by controlling a team of three quadrotors.","keywords":"planning and coordination methods for multiple robots,robot motion; path; task and mission planning and execution,acquisition of planning models for robotics,integrated planning and execution in robotic architectures,real-world planning applications for autonomous robots","bibtex":"@inproceedings {icaps16-98,\r\n track = {Robotics Track},\r\n title = {Real-Time Stochastic Optimal Control for Multi-Agent Quadrotor Systems},\r\n url = {http://www.aaai.org/ocs/index.php/ICAPS/ICAPS16/paper/view/13105},\r\n author = {Vicenç Gómez and Sep Thijssen and Andrew Symington and Stephen Hailes and Bert Kappen},\r\n abstract = {This paper presents a novel method for controlling teams of unmanned aerial vehicles using Stochastic Optimal Control (SOC) theory. The approach consists of a centralized high-level controller that computes optimal state trajectories as velocity sequences, and a platform-specific low-level controller which ensures that these velocity sequences are met. The high-level control task is expressed as a centralized path-integral control problem, for which optimal control computation corresponds to a probabilistic inference problem that can be solved by efficient sampling methods. Through simulation we show that our SOC approach (a) has significant benefits compared to deterministic control and other SOC methods in multi-modal problems with noise-dependent optimal solutions, (b) is capable of controlling a large number of platforms in real-time, and (c) yields collective emergent behavior in the form of flight formations. Finally, we show that our approach works for real platforms, by controlling a team of three quadrotors.},\r\n keywords = {planning and coordination methods for multiple robots,robot motion; path; task and mission planning and execution,acquisition of planning models for robotics,integrated planning and execution in robotic architectures,real-world planning applications for autonomous robots}\r\n}\r\n\r\n","author_short":["Gómez, V.","Thijssen, S.","Symington, A.","Hailes, S.","Kappen, B."],"key":"icaps16-98","id":"icaps16-98","bibbaseid":"gmez-thijssen-symington-hailes-kappen-realtimestochasticoptimalcontrolformultiagentquadrotorsystems","role":"author","urls":{"Paper":"http://www.aaai.org/ocs/index.php/ICAPS/ICAPS16/paper/view/13105"},"keyword":["planning and coordination methods for multiple robots","robot motion; path; task and mission planning and execution","acquisition of planning models for robotics","integrated planning and execution in robotic architectures","real-world planning applications for autonomous robots"],"metadata":{"authorlinks":{}}},"search_terms":["real","time","stochastic","optimal","control","multi","agent","quadrotor","systems","gómez","thijssen","symington","hailes","kappen"],"keywords":["planning and coordination methods for multiple robots","robot motion; path; task and mission planning and execution","acquisition of planning models for robotics","integrated planning and execution in robotic architectures","real-world planning applications for autonomous robots"],"authorIDs":[],"dataSources":["iMkx859KiXcegwsin","EZtZjCTnxcdTTyeij"]}