HOW FAR CAN THE COMPUTER REPLACE THE DRIVER IN AUTOMATIC DRIVING?

This paper considers how far the development of more and more powerful processors and sensors can enable the automatic driving of vehicles. It is based on a paper by H-G Metzler on The Mercedes-Benz VITA II research car. The many tasks of a driver involve many visual senses and a combination of information, decisions and actions. Reasons for automating driving include: (1) improved road safety; (2) increased traffic densities without loss of safety; and (3) more relaxed driving conditions. Video cameras are the only sensors now available, that can provide enough proximity and visual information for automatic driving. Lane recognition is the basis for all other automatic driving functions. The VITA II uses three systems for vehicle recognition. The function of automatic perception and obeying of a traffic sign is especially complex. A central database system is used to integrate sensor signals into a coherent planning and decision making system. Steering is best automated by a steer-by-wire system. Implementing these devices would allow a vehicle to drive at a correct headway distance in a convoy. Three stages of progressive implementation are outlined, that could make fully automatic driving feasible by around the year 2010.

  • Availability:
  • Corporate Authors:

    Institution of Mechanical Engineers

    1 Birdcage Walk
    London SW1H 9JJ,   England 
  • Authors:
    • ASHLEY, C
  • Publication Date: 1995

Language

  • English

Media Info

  • Pagination: p. 18-20
  • Serial:
    • AUTOMOTIVE ENGINEER
    • Volume: 20
    • Issue Number: 5
    • Publisher: Institution of Mechanical Engineers
    • ISSN: 0307-6490

Subject/Index Terms

Filing Info

  • Accession Number: 00716014
  • Record Type: Publication
  • Source Agency: Transport Research Laboratory
  • Files: ITRD
  • Created Date: Jan 31 1996 12:00AM