ABSTRACT
This research deals with an analysis of the driver’s control algorithm as the first step of a modeling of driver. As driver’s tasks are divided into two parts, one is for recognition of the environment around the vehicle, and the other is for the decision of the control output, such as steering torque, throttle and braking control, and so on. On the recognition part, the relative positioning method between driver and road shape is used with radar sight description. On a decision of the control output, the neural network modeling is adopted. And, it is shown that the result of the model agrees with the experimental result. Next, effective factors of all input to the output using the driver model are analyzed. From the results, it is found that a driver always changes his/her main information to decide the control output.
