Safety Assured Trajectory Planning Based on Data-Driven Probabilistic Reachable Set
Pengcheng Wu, Jun Chen · 2024
Chance-constrained optimization can be used to solve motion planning and control problems in the presence of uncertainties. In this paper, we formulate a path planning problem for UAVs with data-driven uncertainties and solve it through a chance-constrained optimization framework considering the convex approximation of the data-driven probabilistic reachable set. In this problem, the uncertainties of the model are converted to a convex approximation of a probabilistic reachable set and are then incorporated into the optimization framework. An efficient convex optimization algorithm is applied to solve the formulated optimization framework, and we achieve an optimal path for the UAV while ensuring the probability of collision is bounded by a threshold. Finally, numerical simulation validates the efficacy of our proposed approach.