Obstacle avoidance for industrial AGVs

The paper deals with the obstacle avoidance problem for Automated Guided Vehicles (AGVs). We propose an automatic algorithm to build new obstacle-free trajectories that leave the original roadmap. The new path has to also comply with the dynamic and kinematic constraints of the AGV. The new paths are generated by means of polar spline curves, lane-change maneuver curves and line segments. The proposed method is validated by means of simulations.