Computation-efficient path planning using evolving artificial repulsive force over expanding obstacles for robotic manipulator.

With growing demand for high productivity in industrial environments, fast path planning for robotic manipulators is critical, as planning time directly impacts production cycle efficiency. Traditional path planning methods, such as grid-based methods, artificial potential field methods, sampling-based methods, and optimization methods, often suffer from high computation costs in high-dimensional space. In particular, artificial potential field methods require the construction of global potential fields or the computation of gradients and attractive forces, increasing the computational burden. To address these issues, a computation-efficient path planning method, using the evolving artificial repulsive force over the expanding obstacles (EARFEO), is proposed. EARFEO incorporates obstacle clustering and virtual obstacle insertion with box-envelope strategy to mitigate the local minima, and iteratively deforms the path using artificial repulsive forces generated by expanding obstacles. A uniform resampling strategy is employed after each expansion step to maintain consistent waypoint distribution, followed by a final refinement phase to ensure complete clearance of the whole path after obstacle expansion is complete. Simulations and experiments across static and dynamic scenarios demonstrate that EARFEO reduces planning time significantly.