Path Planning for Mirror Assembly in Optical Detection Systems: Internal and External Parameter Optimization of Dynamic Movement Primitives

This paper proposes a hierarchical trajectory optimization method for on-orbit assembly of large-aperture optical detection systems using robotic manipulators, aiming to address the threats posed by noncooperative obstacles and mitigate the risks of module interference during assembly. The method integrates dynamic movement primitives to generate an initial trajectory and employs a hybrid optimization framework that combines intrinsic and extrinsic parameters. In the inner layer, the policy improvement with path integrals algorithm is used to optimize the shape parameters of the dynamic movement primitives, enabling the robot end-effector to pass accurately through predefined assembly points while avoiding collisions between assembled and to-be-assembled mirror modules. In the outer layer, model predictive control continuously refines the robot’s motion states, thereby enhancing the system’s ability to respond to changes in noncooperative obstacles and ensuring safe obstacle avoidance. Simulation experiments demonstrate that the proposed method effectively ensures safe motion in dynamic, unstructured space environments and supports successful assembly operations. This study provides a well-structured and scalable path planning solution for intelligent autonomous robotic operations in complex space missions.