Non-parametric inference and coordination for distributed robotics
This paper presents non-parametric methods to infer the state of an environment by distributively controlling robots equipped with sensors. Each robot represents its belief of the environment state with a weighted sample set, which is used to draw likely observations to approximate the gradient of m...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Institute of Electrical and Electronics Engineers (IEEE)
2014
|
| Online Access: | http://hdl.handle.net/1721.1/90614 https://orcid.org/0000-0001-5473-3566 |
| _version_ | 1811094606847148032 |
|---|---|
| author | Julian, Brian John Angermann, Michael Rus, Daniela L. |
| author2 | Lincoln Laboratory |
| author_facet | Lincoln Laboratory Julian, Brian John Angermann, Michael Rus, Daniela L. |
| author_sort | Julian, Brian John |
| collection | MIT |
| description | This paper presents non-parametric methods to infer the state of an environment by distributively controlling robots equipped with sensors. Each robot represents its belief of the environment state with a weighted sample set, which is used to draw likely observations to approximate the gradient of mutual information. The gradient leads to a novel distributed controller that continuously moves the robots to maximize the informativeness of the next joint observation, which is then used to update the weighted sample set via a sequential Bayesian filter. The incorporated non-parametric methods are able to robustly represent the environment state and robots' observations even when they are modeled as continuous-valued random variables having complicated multimodal distributions. In addition, a consensus-based algorithm allows for the distributed approximation of the joint measurement probabilities, where these approximations provably converge to the true probabilities even when the number of robots, the maximum in/out degree, and the network diameter are unknown. The approach is implemented for five quadrotor flying robots deployed over a large outdoor environment, and the results of two separate exploration tasks are discussed. |
| first_indexed | 2024-09-23T16:02:46Z |
| format | Article |
| id | mit-1721.1/90614 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T16:02:46Z |
| publishDate | 2014 |
| publisher | Institute of Electrical and Electronics Engineers (IEEE) |
| record_format | dspace |
| spelling | mit-1721.1/906142022-09-29T17:49:28Z Non-parametric inference and coordination for distributed robotics Julian, Brian John Angermann, Michael Rus, Daniela L. Lincoln Laboratory Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology. School of Engineering Julian, Brian John Rus, Daniela L. This paper presents non-parametric methods to infer the state of an environment by distributively controlling robots equipped with sensors. Each robot represents its belief of the environment state with a weighted sample set, which is used to draw likely observations to approximate the gradient of mutual information. The gradient leads to a novel distributed controller that continuously moves the robots to maximize the informativeness of the next joint observation, which is then used to update the weighted sample set via a sequential Bayesian filter. The incorporated non-parametric methods are able to robustly represent the environment state and robots' observations even when they are modeled as continuous-valued random variables having complicated multimodal distributions. In addition, a consensus-based algorithm allows for the distributed approximation of the joint measurement probabilities, where these approximations provably converge to the true probabilities even when the number of robots, the maximum in/out degree, and the network diameter are unknown. The approach is implemented for five quadrotor flying robots deployed over a large outdoor environment, and the results of two separate exploration tasks are discussed. United States. Air Force (Contract FA8721-05-C-0002) United States. Office of Naval Research (Grant N00014-09-1-1051) National Science Foundation (U.S.) (Grant EFRI-0735953) Lincoln Laboratory Boeing Company 2014-10-07T19:30:20Z 2014-10-07T19:30:20Z 2012-12 Article http://purl.org/eprint/type/ConferencePaper 978-1-4673-2066-5 978-1-4673-2065-8 978-1-4673-2063-4 978-1-4673-2064-1 0743-1546 http://hdl.handle.net/1721.1/90614 Julian, Brian J., Michael Angermann, and Daniela Rus. “Non-Parametric Inference and Coordination for Distributed Robotics.” 2012 51st IEEE Conference on Decision and Control (CDC) (December 2012). https://orcid.org/0000-0001-5473-3566 en_US http://dx.doi.org/10.1109/CDC.2012.6427043 Proceedings of the 2012 51st IEEE Conference on Decision and Control (CDC) Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Institute of Electrical and Electronics Engineers (IEEE) MIT web domain |
| spellingShingle | Julian, Brian John Angermann, Michael Rus, Daniela L. Non-parametric inference and coordination for distributed robotics |
| title | Non-parametric inference and coordination for distributed robotics |
| title_full | Non-parametric inference and coordination for distributed robotics |
| title_fullStr | Non-parametric inference and coordination for distributed robotics |
| title_full_unstemmed | Non-parametric inference and coordination for distributed robotics |
| title_short | Non-parametric inference and coordination for distributed robotics |
| title_sort | non parametric inference and coordination for distributed robotics |
| url | http://hdl.handle.net/1721.1/90614 https://orcid.org/0000-0001-5473-3566 |
| work_keys_str_mv | AT julianbrianjohn nonparametricinferenceandcoordinationfordistributedrobotics AT angermannmichael nonparametricinferenceandcoordinationfordistributedrobotics AT rusdanielal nonparametricinferenceandcoordinationfordistributedrobotics |