Progressive learning for robot impedance control

A new approach to stable learning control by using an excitation scheduling technique is developed, and applied to an impedance learning problem for high-speed robotic assembly. In this paper, a novel technique is developed which guarantees stability by progressively increasing the level of system excitation. The new method termed "progressive learning" uses scheduled excitation inputs that allow the system to learn quasi-static, slow modes in the beginning, followed by the learning of faster modes. As learning progresses, the system is exposed to a broader range of input excitation, which nonetheless does not incur instability and unwanted erratic responses. This new method is presented in the context of high speed robotic assembly, where an impedance control law is learned with this excitation scheduling method. The basic concept of progressive learning is presented first, followed by a detailed description of the algorithm in the context of robotic assembly. Extensive simulation results and their interpretation and discussion are provided. The underpinning theory of progressive learning along with useful guidelines for scheduling the learning procedure are discussed.