| Budget Amount *help |
¥2,300,000 (Direct Cost: ¥2,300,000)
Fiscal Year 1995: ¥500,000 (Direct Cost: ¥500,000)
Fiscal Year 1994: ¥700,000 (Direct Cost: ¥700,000)
Fiscal Year 1993: ¥1,100,000 (Direct Cost: ¥1,100,000)
|
|---|
| Research Abstract |
Recently, many additional systems to improve active safety, stability, and handling have been developed. At an ealy stage of the research for active safety as driver-vehicle-environment system, it is important to know the behavior of man-machine-environment systems. For such purpose, to know the control algorithm of drivers is required.
There are three types of drivers task, one is recognition for the environment, the other one is decision making for the course to follow, and the last one is course tracking with a vehicle input. This research forms the first step in analyzing control algorithm for a driver including these three tasks with neural network system. It is found that the anaytical results with this model well express the control manners of the real driver with the response at time domain. Furthermore, the control algorithm of the driver is analyzed using the sensitivity analysis to the neural network driver model. From the analysis, it is found that the relationship between road shape and the sensitivities is very important at the view point of active safety.
|
