A chaotic neural network as motor path generator for mobile robotics

Abstract

This work aims at developing a motor path generator for applications in mobile robotics based on a chaotic neural network. The computational paradigm inspired by the neural structure of microcircuits located in the human prefrontal cortex is adapted to work in real-time and used to generate the joints trajectories of a lightweight quadruped robot. The recurrent neural network was implemented in Matlab and a software framework was developed to test the performances of the system with the robot dynamic model. Preliminary results demonstrate the capability of the neural controller to learn period signals in a short period of time allowing adaptation during the robot operation.