LEARNING TRANSFER FUNCTION OF A DRIVER USING GENETIC ALGORITHM IN COLLISION AVOIDANCE

Reduction of traffic accidents has become an increasingly important issue during recent years as the advance of the automobile society. Key to coping with this issue has been the investigation of driver behavior, especially in emergent situations such as in the case of sudden encountering of an obstacle during the driving. Assuming an emergent situation of obstacle avoidance, taking full consideration of the nonlinearity of tires, this paper studies the characteristics of the driver under the critical condition by way of genetic algorithm (GA). Computer simulations are conducted to determine that: 1) the GA is found to be a successful way of grasping the driver's characteristics under the defined emergent situation; 2) the initial distance when the driver starts the avoiding operation has great influence on the driver's performance, i.e., a short initial distance makes the driver respond with large gain and small preview time and may necessitate the braking operation, and vice versa; and 3) the changing patterns of the steering gain and preview time gained by GA agree very well with the general empirical knowledge on the driver's characteristics under emergent situations, which verifies the validity of the approach for further applications. With its powerful ability to search optimal solutions for certain problems, GA can be expected to be a very useful method in studying the human factors to provide knowledge for future intelligent traffic systems.

• Supplemental Notes:
  • Five volumes of papers and one volume of abstracts comprise the published set of conference materials.
• Corporate Authors:

  VERTIS

  TORANOMOM 34 MORI BUILDING 1-25-5
  TORANOMON, MINATOKU, TOKYO 105  Japan 
• Authors:
  • NAGAI, M
  • KATAGIRI, T
  • Onda, Masatoshi
• Conference:
  • Steps Forward. Intelligent Transport Systems World Congress
  • Location: Yokohama, Japan
  • Date: 1995-11-9 to 1995-11-11
• Publication Date: 1995-11

Language

• English

Media Info

• Pagination: p. 1767

Subject/Index Terms

• TRT Terms: Behavior; Crash avoidance systems; Driver performance; Drivers; Genetic algorithms; Intelligent transportation systems; Personnel performance; Simulation; Stress (Psychology); Traffic crashes
• Geographic Terms: Japan
• Subject Areas: Highways; Operations and Traffic Management; Safety and Human Factors; I83: Accidents and the Human Factor;

Filing Info

• Accession Number: 00723472
• Record Type: Publication
• Report/Paper Numbers: Volume 4
• Files: TRIS
• Created Date: Jul 29 1996 12:00AM