Trajectory planner for mobile robots using particle swarm optimization

Abstract

The trajectory planning problem is an essential one in robotics. In order to solve this problem, a temporal motion law for a given geometric path has to be defined and a set of requirements for the trajectory has to be fulfilled. The motion planning of a mobile robot must be more than just collision-free: it must be fast, safe and easy to use. For the trajectory planning problem, minimum time to achieve the goal is an important criterion. This paper proposes an original trajectory planning approach using quintic splines in a real environment with several obstacles. The proposed trajectory planning approach is based on the optimization of parameters of the quintic splines using Particle Swarm Optimization. The quintic-splines are ideal since they give continuity in position, heading, curvature, velocity, and acceleration ensuring a smooth movement of the mobile robot. The trajectory planning must provide not only the planning of the path (spatial dimension) but also the speed profile (temporal dimension). The effectiveness of the proposed planner is proved with suitable examples. The experiments show that the proposed planner is able to provide a smooth trajectory.