Flexible Robot Trajectory Tracking Control

Operational problems with space robots relate to several factors, one most importantly being structural flexibility and subsequently significant difficulties with position control. Elastic link vibrations with inherent nonminimum phase response coupled with large rotations create a complex nonlinear dynamic system for control. The ability of the flexible robot to follow a prescribe trajectory is an important issue. In case of more complex operation, good tracking control could potentially save much time and money. This paper compares the performance of control strategies with noise filtering, then combined with nonminimum phase and corrective control action modeled by time delays for a stationary spacecraft mounted flexible robot tracking a 12.6 m × l2.6 m square trajectory. Finally a fuzzy logic system (FLS) vibration suppression strategy demonstrates its effectiveness in providing intelligent and autonomous precision control. The inclusion of this fuzzy logic approach to vibration suppression is motivated by the prospect of substituting for the control of space robotic systems typically teleoperated by astronauts. Tracking results are obtained initially without vibration suppression or noise filtering, then with EKF and FLAEKF state estimators to determine tracking accuracy for robot process and measurement noise. Time delays are included in the feedback and control action loops to simulate nonminimum phase response and corrective control. Finally, a FLS is used to adapt the control law and suppress residual vibrations. Good tracking results were previously achieved using an input shaping method to reduce residual vibrations coupled with an inverse kinematics control strategy and recursive order-n algorithm to model a two-link flexible robot tracking a square trajectory [1].