Oscillation reduction for artificial potential field using vector projections for robotic manipulators

This paper proposes a method of reducing oscillations for integrations of the artificial potential field algorithm in real-time robot manipulator path planning using proximity sensors. Proximity sensing is a technology that has the potential to play an essential role in the development of robotics. It can fulfill the promise of safe, robust, and autonomous systems in industry and everyday life alongside humans by enabling the ability of online and fast-reacting motions without the need for visual mapping. In this context, the artificial potential field technique is one of the most suitable methods for path planning due to its simplicity of application and efficiency in real-time systems without the need for global mapping. Despite its efficiency, this technique is known to be susceptible to problems such as local minima and oscillations within the overall path. To solve this problem, a method of reducing the oscillations is proposed by modifying the direction of the repulsive force to follow its orthogonal projection onto the attractive force. Since with conventional repulsive motions, the robot only moves in one direction, opposite to the obstacles, this approach has better exploitation of the redundancy space to maintain the task motion and is less likely to get stuck in local minima. The effectiveness of this proposed method is demonstrated through simulations on real robot manipulators in comparison with the original artificial potential field technique.