An Intelligent Traction Control for Motorcycles

Abstract

The appearance of anti-lock braking systems (ABS) and traction control systems (TCS) have been some of the most major developments in vehicle safety. These systems have been evolving since their origin, always keeping the same objective, by using increasingly sophisticated algorithms and complex brake and torque control architectures. The aim of this work is to develop and implement a new control model of a traction control system to be installed on a motorcycle, regulating the slip in traction and improving dynamic performance of two-wheeled vehicles. This paper presents a novel traction control algorithm based on the use of Artificial Neural Networks (ANN) and Fuzzy Logic. An ANN is used to estimate the optimal slip of the surface the vehicle is moving on. A fuzzy logic control block, which makes use of the optimal slip provided by the ANN, is developed to control the throttle position. Two control blocks have been tuned. The first control block has been tuned according to the experience of an expert operator. The second one has been optimized using Evolutionary Computation (EC). Simulation shows that the use of EC can improve the fuzzy logic based control algorithm, obtaining better results than those produced with the control tuned only by experience.