Modern road accident investigation and database structures are focused on accident analysis and classification from the point of view of the accident itself. The presented article offers a new approach, which will describe the accident from the point of view of integrated safety vehicle systems. Seven main categories have been defined to specify the level of importance of automated system intervention. One of the proposed categories is a new approach to defining the collision probability of an ego-vehicle with another object. This approach focuses on determining a 2-D reaction space, which describes all possible positions of the vehicle or other moving object in the specified amount of time in the future. This is to be used for defining the probability of the vehicles interacting - when the intersection of two reaction spaces exists, an action has to be taken on the side of ego-vehicle. The currently used 1-D quantity of TTC (time-to-collision) can be superseded by the new reaction space variable. Such new quantity, whose basic idea is described in the article, enables the option of counting not only with necessary braking time, but mitigation by changing direction is then easily feasible. Finally, transparent classification measures of a probable accident are proposed. Their application is highly effective not only during basic accident comparison, but also for an on-board safety system.
Classification Of Road Accidents From The Perspective Of Vehicle Safety Systems
2015
Article (Journal)
Electronic Resource
Unknown
Metadata by DOAJ is licensed under CC BY-SA 1.0
Data mining on road safety: factor assessment on vehicle accidents using classification models
Taylor & Francis Verlag | 2016
|Vehicle characteristics and road accidents
Engineering Index Backfile | 1954
|Effective Active Safety to Reduce Road Accidents
British Library Conference Proceedings | 1995
|Effective active safety to reduce road accidents
Automotive engineering | 1995
|