Reactive motion planning with probabilisticsafety guarantees

Yuxiao Chen, Ugo Rosolia, Chuchu Fan, Aaron Ames, Richard Murray
Proceedings of the 2020 Conference on Robot Learning, PMLR 155:1958-1970, 2021.

Abstract

Motion planning in environments with multiple agents is critical to many important autonomous applications such as autonomous vehicles and assistive robots. This paper considers the problem of motion planning, where the controlled agent shares the environment with multiple uncontrolled agents. First, a predictive model of the uncontrolled agents is trained to predict all possible trajectories within a short horizon based on the scenario. The prediction is then fed to a motion planning module based on model predictive control. We proved generalization bound for the predictive model using three different methods, post-bloating, support vector machine (SVM), and conformal analysis, all capable of generating stochastic guarantees of the correctness of the predictor. The proposed approach is demonstrated in simulation in a scenario emulating autonomous highway driving.

Cite this Paper

BibTeX
@InProceedings{pmlr-v155-chen21e, title = {Reactive motion planning with probabilisticsafety guarantees}, author = {Chen, Yuxiao and Rosolia, Ugo and Fan, Chuchu and Ames, Aaron and Murray, Richard}, booktitle = {Proceedings of the 2020 Conference on Robot Learning}, pages = {1958--1970}, year = {2021}, editor = {Kober, Jens and Ramos, Fabio and Tomlin, Claire}, volume = {155}, series = {Proceedings of Machine Learning Research}, month = {16--18 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v155/chen21e/chen21e.pdf}, url = {https://proceedings.mlr.press/v155/chen21e.html}, abstract = {Motion planning in environments with multiple agents is critical to many important autonomous applications such as autonomous vehicles and assistive robots. This paper considers the problem of motion planning, where the controlled agent shares the environment with multiple uncontrolled agents. First, a predictive model of the uncontrolled agents is trained to predict all possible trajectories within a short horizon based on the scenario. The prediction is then fed to a motion planning module based on model predictive control. We proved generalization bound for the predictive model using three different methods, post-bloating, support vector machine (SVM), and conformal analysis, all capable of generating stochastic guarantees of the correctness of the predictor. The proposed approach is demonstrated in simulation in a scenario emulating autonomous highway driving.} }
Endnote
%0 Conference Paper %T Reactive motion planning with probabilisticsafety guarantees %A Yuxiao Chen %A Ugo Rosolia %A Chuchu Fan %A Aaron Ames %A Richard Murray %B Proceedings of the 2020 Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2021 %E Jens Kober %E Fabio Ramos %E Claire Tomlin %F pmlr-v155-chen21e %I PMLR %P 1958--1970 %U https://proceedings.mlr.press/v155/chen21e.html %V 155 %X Motion planning in environments with multiple agents is critical to many important autonomous applications such as autonomous vehicles and assistive robots. This paper considers the problem of motion planning, where the controlled agent shares the environment with multiple uncontrolled agents. First, a predictive model of the uncontrolled agents is trained to predict all possible trajectories within a short horizon based on the scenario. The prediction is then fed to a motion planning module based on model predictive control. We proved generalization bound for the predictive model using three different methods, post-bloating, support vector machine (SVM), and conformal analysis, all capable of generating stochastic guarantees of the correctness of the predictor. The proposed approach is demonstrated in simulation in a scenario emulating autonomous highway driving.
APA
Chen, Y., Rosolia, U., Fan, C., Ames, A. & Murray, R.. (2021). Reactive motion planning with probabilisticsafety guarantees. Proceedings of the 2020 Conference on Robot Learning, in Proceedings of Machine Learning Research 155:1958-1970 Available from https://proceedings.mlr.press/v155/chen21e.html.

Related Material