Automatic driving risk evaluation method considering uncertainty

Abstract

The invention relates to an automatic driving risk evaluation method considering uncertainty. The automatic driving risk evaluation method considering uncertainty comprises the steps of a priori environment map constructing step, a traffic participation vehicle state observing step, an own vehicle executing trajectory taking step and a self-adaptive multi-target tracking step based on priori knowledge. Through driving behavior likelihood estimation, orbifold matching and trajectory predication, running collision risk is estimated by using multiple safety margins, and finally, space-time running risk situation map is output for evaluation; and therefore, multi-safety-margin risk estimation of automatic driving can be realized, the adapting ability of automatic driving to the environment canbe improved, and the automatic driving process is more stable and reliable.

Description

technical field [0001] The invention relates to the technical field of risk assessment of smart cars, in particular to a risk assessment method for automatic driving in consideration of uncertainty. Background technique [0002] With the rapid development of society and economy, there are more and more motor vehicles on the road, and the problem of driving safety has become more and more prominent. Unmanned driving technology, which is expected to alleviate traffic pressure and drive safety, is gradually moving from the laboratory to the public. High requirements for driving safety require self-driving vehicles to have the ability to detect risks before humans and deal with risks before humans, so as to avoid collisions as much as possible and minimize damage. However, due to the limitations of sensor resolution and measurement accuracy, the unobservable behavior of traffic-involved drivers and the difficulty in measuring the parameters of traffic-involved vehicles, the envi...

AI Technical Summary

Problems solved by technology

However, due to the limitation of sensor resolution and measurement accuracy, the unobservable behavior of traffic-involved drivers and the difficulty in measuring the parameters of traffic-involved vehicles, the environmental perception results of smart vehicles inevitably have high uncertainty.

Images