Nonlinear Optimization with Singularity Map for Space Robot Trajectory Planning

Abstract

This Master's thesis conducted at the DLR's Robotik und Mechatronik Zentrum in Oberpfaffenhofen (Munich, Germany) introduces a methodology for optimal trajectory planning in the context of On-Orbit Servicing (OOS), a rapidly expanding field within military, commercial, and civil satellite sectors. The complexity of optimal trajectory planning for a free-floating robot is inherently non-convex and highly constrained due to the robot’s free-floating nature, which makes the avoidance of dynamic singularities a significant challenge. The research proposes incorporating statically calculated singular configurations, as an inequality constraint within the trajectory optimization. This approach ensures that computed trajectories maintain a threshold distance from singular configurations. By demonstrating its efficacy for a 4DOF robotic system, the study shows that a Singularity map can be successfully integrated within an optimization framework to avoid Singularities during motion planning.