1. Home
  2. TRID
  3. View Record
https://nap.nationalacademies.org/catalog/27432/critical-issues-in-transportation-for-2024-and-beyond

Quantitative Test for FCWS of Transit Buses

With the development of remote sensor technology and its application, frontal collision warning system for vehicle become one of the most active areas in safety related research. It is generally accepted that this technology composes the following parts: (a) Remote sensing for target detection and target tracking; (b) Target state estimation: To determine if the object is static or moving. If moving, its position, velocity and acceleration if possible. In general, the more information available, the easier for decision making; (c) Threat assessment: According to the subject vehicle position, velocity, and acceleration to the object(s) detected to decide any threat to the subject vehicle and the degree of the threat; There are many algorithms available for threat assessment [ARAKI][PENG2]. From vehicle kinematics model, it would be ideal to involve relative acceleration in threat assessment. However, unless there are wireless communication between vehicles, acceleration measured and/or estimated from remote sensors generally subject to two major limits: large error and time delay; Some are speed based in the sense that acceleration are chosen as some nominal values. (d) Warning issuance if threat degree is over some specified threshold: (e) Display to the driver the warning signal to the driver in a smart and sensible way considering human factor; One of the important problems in the development of collision warning system is how to evaluate the system’s performance. There are different approaches which may be considered complementary to each other. [PENG1] and [PENG2] use naturalistic driving data set for the evaluation of several threat assessment algorithms. Work in [YANG] use an analytical method to compare different threat assessment algorithms to evaluate its effectiveness under parameter variation. The third approach is to physically test the system under proper conditions. The same as other physical systems, Frontal Collision Warning Systems (FCWS) for transit bus needs extensive test in a known environment which is similar to urban and suburban highway and drive way. Test of FCWS was also conducted by CAMP project as reported in [KIEF1]. A specific maneuver related test (driver last minute braking) test is conducted for drivers in different age group and reported in [KIEF2], which can be described as Human Factor related. However, CAMP project is mainly for evaluation purposes. Our test, on the other hand, is for two purposes: (a) to preliminarily evaluate the performance of the FCWS algorithm developed at California PATH; (b) To test the measurement and estimation error characteristics of the environment based on vehicle on-board sensors. The test data can be used for system evaluation, algorithm tuning and further development purposes. In any urban and suburban driving environment, objects or hazards in bus drive way can be divided into two categories: moving objects and static objects. The test environment is created purposely and thus known in the sense the following senses: Moving object - its velocity and position with respect to an inertial coordinate system are recorded in realtime; Static object - its position is also recorded. If the subject vehicle moves in a specified manner from a known initial position, then its moving history is known at any time. In this way, a known inter-relationship between the subject vehicle and the environment is created. Due to the phase of the project, the test conducted is restricted to vehicle moving straight ahead instead of on curved road. However, similar tests can be conducted for any other environment in future development, for example in curved road, or up/down a hill. Test site is chosen at Crows Landing, an abandoned NASA airfield, which provides multiple straight lanes (runways).

  • Corporate Authors:

    ITS America

    1100 17th Street, NW, 12th Floor
    Washington, DC  United States  20036
  • Authors:
    • Lu, Xiao-Yun
    • Zhang, Yong-Quan
    • Marco, David
    • Chang, Joanne
    • Johnston, Scott
    • Zhang, Wei-Bin
    • Shladover, Steven
  • Conference:
  • Publication Date: 2005

Language

  • English

Media Info

  • Media Type: Print
  • Features: CD-ROM; Figures; References;
  • Pagination: 11p
  • Monograph Title: Proceedings of the 12th World Congress on Intelligent Transport Systems

Subject/Index Terms

Filing Info

  • Accession Number: 01015833
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Jan 13 2006 10:58AM