AN UNSUPERVISED NEURAL-NETWORK FOR LOW-LEVEL CONTROL OF A WHEELED MOBILE ROBOT - NOISE RESISTANCE, STABILITY, AND HARDWARE IMPLEMENTATION

We have recently introduced a neural network mobile robot controller ( NETMORC). This controller, based on previously developed neural networ k models of biological sensory-motor control, autonomously learns the forward and inverse odometry of a differential drive robot through an unsupervised learning-by-doing cycle. After an initial learning phase, the controller ran move the robot to an arbitrary stationary or movin g target while compensating for noise and other forms of disturbance, such as wheel slippage or changes in the robot's plant. In addition, t he forward odometric map allows the robot to reach targets in the abse nce of sensory feedback. The controller is also able to adapt in respo nse to long-term changes in the robot's plant, such as a change in the radius of the wheels. In this article we review the NETMORC architect ure and describe its simplified algorithmic implementation, we present new, quantitative results on NETMORC's performance and adaptability u nder noise-free and noisy conditions, we compare NETMORC's performance on a trajectory-following task with the performance of an alternative controller, and we describe preliminary results on the hardware imple mentation of NETMORC with the mobile robot ROBUTER.