Suggests a new approach to man-machine interfaces that is based on the characteristics of an imaginary individual's driving behavior and easily detected car control figures. An adaptive resonance theory (ART) network is combined with fuzzy neural network (FNN) learning to obtain safe car-following headways between cars that are traveling at different speeds. The framework of this system is divided into two major parts. The first part uses the ART network, which uses the difference between the required least horizontal braking distance of a car and the distance between the car and the obstacle standing at the end of the lane, at the moment when the driver applies the brakes, during a test where the car is traveling on a straight lane at a high speed. Off-line self-organizing cluster discovery is then conducted. The clusters of the two distances obtained through the ART network are used to determine the levels of the risk-taking factors of the drivers. Then the output of this ART network, the car speed and the free travel of the brake pedal are used as the FNN's input. The headway is the FNN's output. Through the learning capability of neural nets, the complex membership functions between the inputs and the output can be established. Finally, the appropriate headways predicted by the FNN are compared with actual field data to prove the accuracy of the system in predicting the different safe headways, given different driver and car characteristics at different speeds. The completion of this system provides a feasible course for the development of a neural net in an individually-oriented man-machine interface system.
Car cruising system incorporating ART-FNN man-machine interface in adaptation to various driving behaviors
International Journal of Computer Applications in Technology ; 12 , 2-5 ; 160-173
1999
14 Seiten
Article (Journal)
English
European Patent Office | 2018
|SAE Technical Papers | 2003
|TIBKAT | 1977
|