A novel pose control framework and its implementation for robot manipulators following constrained spatial paths

This paper develops a novel pose control framework for robot manipulators traversing a given spatial curve with constant speed. The key to this framework is the use of a Rotation Minimizing Frame (RMF) for path generation and control, enhancing motion stability for paths with significant curvature and inflection points, and reducing kinematic twist. Using the governing equations based on the RMF, we first develop the reference velocity and acceleration along the path that is consistent with the RMF. Employing tools from differential geometry, we derive a path following position control law by projecting the robot translation states onto the RMF. From an analytical description of the relative orientation error kinematics, we derive a stabilizing orientation controller by utilizing the Modified Rodrigues Parameters to avoid the unwinding problem. The proposed framework is applicable to both torque-controlled and velocity-controlled robots, and we provide results from real-time experiments on both types of robots to verify the effectiveness and advantages of the proposed approach.