Developing adaptive driving route guidance systems based on fuzzy neural network

With the advent of communication and information technologies, much information about route conditions and driver behaviours can be acquired by the mobile devices and used to guide drivers while driving vehicles to the destination. However, even though with the same starting and destination points, different drivers typically do not go through the same path as they make choices of routes based on different attributes and weighting of the attributes. Driving route guidance systems need to be able to adapt to various behaviours of drivers to generate routes teller-made for individual drivers. The perceptions of the drivers being fed back to the system are linguistic values which are vague. A neural fuzzy approach is used to learn the decision logic with vague attributes from the past driving records of the drivers to make the guidance system adaptive. The adaptive-network-based fuzzy inference system (ANFIS) is used so that the system can self-adjust without user intervention. By integrating this intelligent adaptive capability, a driving route guidance system is proposed in this paper which is capable of adapting to individual drivers gradually to provide different optimum routes automatically based on different preferences.