An adaptive neural controller for a tendon driven robotic hand

In this paper we present our ongoing work on the control of a tendon driven robotic hand by an adaptive learning mechanism evolved using a simulator developed over the last years. We present the “ligand-receptor” concept that can be easily used by artificial evolution to explore (a) the growing of a neural network, (b) value systems and (c) learning mechanisms systematically for a given task (i.e., grasping). The proposed neural network allows the robotic hand to explore its own movement capabilities to interact with objects of different shape, size and material and learn how to grasp them. As the evolved neural controller is highly adaptive, it will allow us in the future to systematically investigate the interplay between morphology and behavior using the same, but adaptive neural controller.