Low Computational Cost for Multiple Waypoints Trajectory Planning: A Time-Optimal-Based Approach

Abstract

In the field of mobile robots, achieving minimum time in executing trajectories is crucial for applications like delivery, inspection, and search and rescue. In this article, a novel time-optimal planner based on optimization methods is introduced. Despite the high computational cost associated with such methods, the solution calculates time-optimal multi-waypoint trajectories, achieving results in the order of milliseconds. The proposed method formulates a time-optimal trajectory using the Pontryagin's maximum principle as a policy. By utilizing a point mass model, the planner generates trajectories that are adaptable to different robot models. The approach incorporates a definition of a search space to guarantee convergence while considering the system limits. Simulation and real-world experiments are performed to validate the feasibility of our method with different configurations. Simulation results compared to a benchmark method demonstrate our approach's superior performance in terms of computational time, achieving near-optimal solutions. In addition, in the real-world experiments, the integration of the method into practical applications is validated.