Obstacle Avoidance Method for On-orbit Assembly Based on Artificial Potential Field and Improved A* Path-planning Algorithm

This paper presents a dynamic modeling and control strategy for on-orbit assembly of small space modules. For attitude dynamics control, feedback linearization is applied to decouple nonlinear characteristics, while a proportional-differential (PD) control law with compensation terms is developed for relative translational dynamics control, transforming the original systems into canonical second-order systems. Control parameters are selected according to predefined performance indices to satisfy specified time-domain response requirements. Furthermore, to address collision avoidance during on-orbit assembly, artificial potential field (APF) method is incorporated into the baseline control law, and Hessian matrix is applied to identify local minima of APFs. An improved A* algorithm with prioritized node exploration order is designed to assist escape local minima, reducing cost ambiguity under three-dimensional (3D) astrodynamical scenario. Simulations verify the functionality of the path-planning algorithm, and the optimal step size of path node selection is determined through efficiency metrics of multiple simulations.