Design of a digital adaptive algorithm for torque-controlled robotic joints

Publisher Summary High-performance motion and force control of robot manipulators rely on the ability of the actuation system to provide accurate and reliable control of joint torques. This ability is severely limited in conventional robots, and a number of efforts are directed toward the development of torque sensors, actuator–transmission systems, and joint torque controllers. This chapter presents the development of an adaptive joint torque control system for a prototype link of ARTISAN, a ten-degree-of-freedom manipulator currently under development at Stanford University. Due to nonlinear effects and parameter variations, a simple fixed controller is insufficient to provide robust control over various configurations. The chapter discusses the design of a pole-placement based joint torque controller and employs a model reference adaptive control (MRAC) scheme to provide more robust control over variations. A model of the actuator–transmission system, including flexibilities, is developed and simulation results of the proposed control systems are presented. The simulation results demonstrate the effectiveness of the controller in maintaining high performance over a wide range of parameter uncertainties.