C-OPT: Coverage-Aware Trajectory Optimization Under Uncertainty

Abstract

We introduce a new problem of continuous, coverage-aware trajectory optimization under localization and sensing uncertainty. In this problem, the goal is to plan a path from a start state to a goal state that maximizes the coverage of a user-specified region while minimizing the control costs of the robot and the probability of collision with the environment. We present a principled method for quantifying the coverage sensing uncertainty of the robot. We use this sensing uncertainty along with the uncertainty in robot localization to develop C-OPT, a coverage-optimization algorithm which optimizes trajectories over belief-space to find locally optimal coverage paths. We highlight the applicability of our approach in multiple simulated scenarios inspired by surveillance, UAV crop analysis, and search-and-rescue tasks. We also present a case study on a physical, differential-drive robot. We also provide quantitative and qualitative analysis of the paths generated by our approach.