Intelligent control for autonomous vehicles using real-time adaptive associative memory neural networks

Addresses the problem of adaptively controlling an autonomous vehicle. The plant is a complex, nonlinear function of many parameters, some of which will be time varying (e.g. vehicle mass), and operating in a dynamic environment (e.g. varying tyre/road friction coefficient). A priori modelling is a very time consuming and complex process, so a real-time, nonlinear adaptive algorithm is required which, for safety reasons, must have an initial rapid convergence rate and guaranteed long term convergence. The neuronally inspired Albus CMAC and adaptive B-splines have previously been identified as possessing these properties. The algorithms, their implementation cost and the training rules are described in this paper, as well as discussing the similarities between these algorithms and fuzzy logic.