End-Effector Trajectory Optimization and Obstacle Avoidance for Free-Flying Space Robots

This paper investigates the Admissible Subspace Trajectory Optimizer (ASTRO) algorithm and its application to spacecraft-mounted robotic manipulators for end-effector obstacle avoidance and trajectory optimization. ASTRO is a real-time path planning algorithm that prioritizes finding an admissible trajectory over optimality, ensuring feasible trajectories under strict time limitations. To represent the robotic manipulator's physical limits within the algorithm, the workspace is represented using two convex geometric constraints: an inner constraint that repels the end-effector path; and an outer constraint that attracts it. Combined with the algorithm's ability to represent environmental obstacles as geometric constraints, this approach produces effective results in path planning and obstacle avoidance. When given a longer computation time, the results show that the algorithm is also able to find the optimal solution. The results are produced in a dynamically accurate offline planning environment and are experimentally validated on the Spacecraft Proximity Operations Testbed at Carleton University.