A new local path planner for a nonholonomic wheeled mobile robot in cluttered environments

This paper presents a new local path planner based on distance information, for mobile robots with nonholonomic constraints. The nearby obstacles are mapped as linear constraints over the robot's velocities to form a Feasible Velocities Polygon. This Polygon represents the set of velocities that the robot can use without collision with the obstacles, The planner, composed by two modules, uses the FVP representation to ensure the collision-free navigation. The first module allows the robot to continuously approach the goal position, avoiding the obstacles and following a stable reference trajectory, obtained from an exponential control law. When a deadlock situation occurs, the second module allows the robot to follow the obstacle's boundary in order to escape the deadlock. The presented results demonstrate the capabilities of the proposed method for solving the collision-free path-planning problem.