Kinematic motor learning

This paper focuses on adaptive motor control in the kinematic domain. Several motor learning strategies from the literature are adopted to kinematic problems: “feedback-error learning”, “distal supervised learning”, and “direct inverse modeling”. One of these learning strategies, direct inverse modeling, is significantly enhanced by combining it with abstract recurrent neural networks. Moreover, a newly developed learning strategy (“learning by averaging”) is presented in detail. The performance of these learning strategies is compared on different learning tasks on two simulated robot setups (a robot-camera-head and a planar arm). The results indicate a general superiority of direct inverse modeling if combined with abstract recurrent neural networks. Learning by averaging is especially successful if the motor task is constrained by special requirements.