This academic article introduces a new trajectory optimization strategy for mobile robots navigating in complex dynamic environments. The method generates multiple distinct evasive trajectories, which are then optimized by locally constrained parallel planners. Each planner attempts a different evasive maneuver, and the robot executes the feasible trajectory with the lowest cost. This approach improves the robot’s efficiency and safety by avoiding locally optimal solutions and unnecessary trajectory switches, which can lead to unsafe robot motion. The authors demonstrate their method’s effectiveness with a mobile robot navigating among pedestrians.
Publication date: 12 Jan 2024
Project Page: Not provided
Paper: https://arxiv.org/pdf/2401.06021
