A local planner for Ackermann-driven vehicles in ROS SBPL

Abstract

For a mobile service robot, safe navigation is essential. To do so, the robot needs to be equipped with collision avoidance and global path planning capabilities. The target platforms considered in this paper are Ackermann-driven vehicles. Hence, a planning approach which directly takes the kinematic and dynamic constraints of the vehicle into account is required. The Search-based Planning Library (SBPL) package of the Robot Operating System (ROS) provides global path planning which takes these constraints into account. However, it misses a local planner that can also make use of the Ackermann kinematic constraints for collision avoidance. A local planner is useful to take dynamic obstacles into account as early as possible. In this paper, we extend the SBPL included in ROS by a local planner, which makes use of motion primitives. We extend the ROS package and show first experimental results on Ackermann-driven vehicles.