Driver misoperation risk early warning method and system

Abstract

The invention discloses a driver misoperation risk early warning method and system, and belongs to the technical field of intelligent driving; the method comprises the steps: obtaining a real-time driving video of a to-be-detected driver as to-be-detected data; analyzing the to-be-measured data to obtain a prediction variable, and converting the prediction variable into a corresponding prediction vector; and taking the prediction vector as an input of a pre-trained risk early warning model to obtain a risk early warning result of the to-be-tested driver. According to the method, the influence of cognitive load on misoperation is considered, the action of action coordination degree and negative emotion variables is also added, risk factors of the misoperation are considered more completely, and the risk prediction capability and prediction accuracy of the misoperation are greatly improved.

Description

technical field [0001] The invention relates to the technical field of intelligent driving, in particular to a risk warning method and system for a driver's misoperation. Background technique [0002] At present, the driver's driving state is generally judged by the driver's performance during the driving process. The cognitive load detection methods in the existing technology mainly include the following types: [0003] (1) Subjective measurement method: the subjects are required to state their own cognitive load and make a self-evaluation report on their own experience state; [0004] (2) Task performance measurement method: through the task performance, behavior records, and test scores obtained by the subjects under different cognitive load conditions, the degree of cognitive load can be inferred from the level of performance and performance; [0005] (3) Physiological correlation test method: By establishing the correlation between physiological signals and cognitive l...

AI Technical Summary

Problems solved by technology

[0011] The defective that above-mentioned prior art exists is: original technical limitation is:

[0012] (1) The original technology did not distinguish the two concepts of cognitive load and misoperation. From the perspective of cognitive neuroscience and ergonomics, high cognitive load does not mean that misoperation must occur. Influencing factors also include negative emotions, attentional bias, and other factors. Therefore, when designing and developing early warning equipment for misoperation, other factors that affect misoperation should be further included in the early warning predictor variables when considering cognitive load detection;

[0013] (2) Many technologies in the original technology have not been considered applicable to the driving environment and driving scene. Place large sensing and computing devices inside;

[0014] (3) Many technologies in the original technology have not been tested in real driving scenes or simulation scenes specially oriented to driving characteristics, and experimental data have been obtained, and early warning models have been designed based on these experimental data, which lacks scientific and effective proof

Images