SYSYOCJan 10, 2021

Using Uncertainty Data in Chance-Constrained Trajectory Planning

arXiv:1904.128259 citationsh-index: 34
Originality Incremental advance
AI Analysis

For autonomous systems operating in dynamic environments, this provides a theoretically grounded method for chance-constrained planning with learned uncertainties.

The paper tackles trajectory planning with obstacles of uncertain location, where uncertainty distributions are learned online. It derives tight concentration bounds for moment estimation and a tractable mixed-integer convex reformulation, achieving high-confidence feasibility.

We consider the problem of trajectory planning in an environment comprised of a set of obstacles with uncertain locations. While previous approaches model the uncertainties with a prescribed Gaussian distribution, we consider the realistic case in which the distribution's moments are unknown and are learned online. We derive tight concentration bounds on the error of the estimated moments. These bounds are then used to derive a tractable and tight mixed-integer convex reformulation of the trajectory planning problem, assuming linear dynamics and polyhedral constraints. The solution of the resulting optimization program is a feasible solution for the original problem with high confidence. We illustrate the approach with a case study from autonomous driving.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes