Theory of collision avoidance capability in automated driving technologies

Paper #:
  • 2018-01-0044

Published:
  • 2018-04-03
Abstract:
The collision avoidance capability of automated vehicles consists of response capability and preparation capability. A collision occurs when the a-th vehicle's path and the b-th vehicle path have a crossing point and the two vehicles drive on the crossing point at same time. To describe the above conditions, we introduce candidate path matrix R_{an} representing the a-th vehicle's path with dense way-points referred by n and conflict tensor C_{an,bm} representing whether there is enough arrival time interval to avoid collision with the b-th vehicle or not. Then the collision avoidance equation of the automated vehicle a is given by a following equation in terms of R_{an} and its speed v_a, Σ_{nm} R_{an} C_{an,bm}(v_a,v_b) R^n_{bm}= 0 where the b-th vehicle condidate matrix R^n_{bm} and its speed v_b are evaluated with "normal driving" model. On the concise representation, we consider a behavioral hazard describing the candidate path matrix change from R^n_{bm} to R^h_{bm}. If we replace R^n_{bm} with (R^n_{bm}^{n}+R^h_{bm}) in the equation, its solution R_{an} grantees no collision with the b-th vehicle including the behavioral hazard. Solving the equation equals to that the automated vehicle has prepared for the behavioral hazard beforehand. There are two ways in collision avoidance. One is "response" that the automated vehicle begins its maneuver to avoid collision when a hazard is detected. Another is the above "preparation". We conclude that the collision avoidance capability consists of response capability supporting accuracy of collision avoidance after extreme conditions and preparation capability supporting accuracy to avoid that it goes to extreme conditions. The preparation capability is evaluated through stability analysis of the automated vehicle behaviors.
Access
Now
SAE MOBILUS Subscriber? You may already have access.
Buy
Attention: This item is not yet published. Pre-Order to be notified, via email, when it becomes available.
Select
Price
List
Download
$22.00
Mail
$22.00
Members save up to 36% off list price.
Share
HTML for Linking to Page
Page URL

Related Items

Technical Paper / Journal Article
1981-09-01
Article
2017-07-26
Technical Paper / Journal Article
1990-02-01
Training / Education
2010-08-03
Training / Education
2005-11-15
Technical Paper / Journal Article
1990-02-01