Dynamic Programming Approach for Motion Planning with Arrival Requirements

Abstract

Motion planning with predictable timing and velocity is crucial in certain tasks in multi-robot coordination, such as role assignment and formation positioning in robot soccer and autonomous intersection management. This paper introduces a dynamic programming algorithm that is polynomial-time in finding the set of feasible setpoint schedules for a mobile robot to reach a specific position on a trajectory at a specific time and a specific velocity. Although the running time of the algorithm is longer than that of the bisection method we proposed previously, the amortized cost of plan generations is much smaller, making it suitable for real time update of setpoint schedules to cope with control and sensing errors.