Fuzzy modeling control for robotic gait synthesis

Abstract

This paper presents a biped locomotion learning scheme using a fuzzy modeling neural network. The learning scheme can generate walking gaits by providing a reference trajectory which defines the desired step width, height and period in several stages. This proposed scheme uses a fuzzy controller combined with a linearized inverse biped model. A multilayer fuzzy neural network is used as a controller; it provides the control signals in each stage of a walking gait. The algorithm used to train the network is the backpropagation through time. The linearized inverse biped model provides the error signals which can be used to back propagate through the controller in each stage. The simulation results are described for a five-link biped walking robot.