Online driving danger estimating method based on offline vehicle deceleration curve

Abstract

The invention discloses an online driving danger estimating method based on an offline vehicle deceleration curve. The method comprises the steps of 1), acquiring a near accident vehicle decelerationcurve sample in natural driving researching as an offline risk vehicle deceleration database, through clustering analysis of the deceleration curves in the database, obtaining different grade kinds ofthe driving risk state at a vehicle braking collision preventing starting time; and 2), based on the offline risk vehicle deceleration database, performing risk state dividing fuzzy rule extraction on the danger estimation index at a sample braking collision preventing starting time according to the offline driving risk state kind result, and obtaining an online driving risk state dividing methodbased on a danger estimation index fuzzy rule. The online driving danger estimating method prevents a restriction problem which may exists in a single danger estimation index and furthermore reflectsthe time when a driver thinks inevitable collision, thereby improving effectiveness and user acceptance of an anti-collision system.

Description

technical field [0001] The invention relates to the technical field of traffic safety evaluation and intelligent traffic system active safety, in particular to an online driving risk estimation method based on an offline vehicle deceleration curve. Background technique [0002] Generally speaking, the degree of criticality (or conflict severity) of a certain situation during driving can be determined by dividing indicators based on the initial conflict state, that is, calculating the risk estimation indicator by obtaining the vehicle motion parameters at the initial moment of the situation (such as based on distance and Time parameters of the final inter-vehicle distance PICUD under emergency braking, collision time TTC and inter-vehicle time THW, etc., are also often referred to as driving safety indicators), and by comparing the calculated value of the risk estimation index with the different preset thresholds of the index (representing The size of different risk levels) d...

AI Technical Summary

Problems solved by technology

This kind of risk estimation index describing the initial conflict state is mainly used for online real-time risk level division, and plays a core role in the anti-collision system. However, the difficulty lies in the determination of the thresholds of different risk level indicators. There is no recognized and perfect method in the literature at present. Classification method of online risk level

Images