Robotic Path Planning Strategy Based on Improved Artificial Potential Field

摘要

The artificial potential field (APF) algorithm, proposed by Khatib, is frequently utilized in the pathfinding and obstacle avoidance projection of moveable robots with known destination. Aiming at the confinements of the traditional APF technique, which are defined as local minimum and unreachable problem respectively, based on the thread of disturbing the mechanical model of force balance to break the equilibrium state, the following improved algorithm is proposed. By sensing the variation of the distance between the robot and the target in certain time unit, the algorithm can quantitate the tendency of robot to be trapped in the mechanical equilibrium state in the artificial potential field. An optimization and improvement of the repulsive force pointing angle and the gravitational field function has been done so that the robot can take the initiative to detach from the state of mechanical equilibrium induced by obstacles. In the MATLAB simulation, the algorithm demonstrates its efficiency and universality compared with traditional algorithm.