REVISE: Robust Probabilistic Motion Planning in a Gaussian Random Field

Alex Rose1, 3, Naman Aggarwal1, Christopher Jewison2, Jonathan How1
1Massachusetts Institute of Technology 2Draper 3Draper Scholar
arXiv Code Documentation
Trajectories generated by REVISE and a baseline covariance steering algorithm in an uncertain wind field. REVISE navigates around a high-variance area and has lower state error.

Simulated trajectories following a control plan generated by a baseline covariance steering algorithm (left) and by REVISE (right) in an uncertain wind field. (Top) REVISE finds a path that mostly avoids a high-variance area (dark green), while the baseline algorithm plans a path which goes directly through the high-variance area. REVISE also achieves lower covariance at the goal (visualized by the size of the yellow goal ellipse). (Bottom) REVISE consistently achieves lower position error at the final state than the baseline method.

Abstract

This paper presents Robust samplE-based coVarIance StEering (REVISE), a multi-query algorithm that generates robust belief roadmaps for dynamic systems navigating through spatially dependent disturbances modeled as a Gaussian random field. Our proposed method develops a novel robust sample-based covariance steering edge controller to safely steer a robot between state distributions, satisfying state constraints along the trajectory. Our proposed approach also incorporates an edge rewiring step into the belief roadmap construction process, which provably improves the coverage of the belief roadmap. When compared to state-of-the-art methods, REVISE improves median plan accuracy (as measured by Wasserstein distance between the actual and planned final state distribution) by 10x in multi-query planning and reduces median plan cost (as measured by the largest eigenvalue of the planned state covariance at the goal) by 2.5x in single-query planning for a 6DoF system.

BibTeX


	@misc{rose2024reviserobustprobabilisticmotion,
	      title={REVISE: Robust Probabilistic Motion Planning in a Gaussian Random Field}, 
	      author={Alex Rose and Naman Aggarwal and Christopher Jewison and Jonathan P. How},
	      year={2024},
	      eprint={2411.13369},
	      archivePrefix={arXiv},
	      primaryClass={cs.RO},
	      url={https://arxiv.org/abs/2411.13369}, 
	}