NEURAL-NETWORK MBPC FOR MOBILE ROBOT PATH TRACKING

This paper presents a way of implementing a model-based predictive con troller (MBPC) for mobile robot path tracking. The method uses a non-l inear model of mobile robot dynamics and thus allows an accurate predi ction of the future trajectories. Constraints on the maximum attainabl e speeds are also considered by the algorithm. A multilayer perceptron is used to implement the MBPC. The perceptron has been trained to rep roduce the MBPC behaviour in a supervised way. Experimental results ob tained when applying the neural network controller to a TRC labmate mo bile platform are given in the paper.