Intelligent vehicle drivability automatic evaluation method fusing multi-source vehicle-mounted sensor information

Abstract

The invention provides an intelligent vehicle drivability automatic evaluation method fusing multi-source vehicle-mounted sensor information. The method comprises the following steps: synchronously obtaining a real vehicle signal necessary for drivability evaluation through an intelligent vehicle test platform; processing a real vehicle signal by adopting a time-frequency analysis method and an abnormal point elimination method, and then completing information fusion on a target layer and a data layer to obtain a data set; analyzing the data set in combination with driving condition characteristics, and intercepting a specific condition through a sliding window method; finding main influence factors of the drivability of the intelligent vehicle by adopting a principal component analysis method, extracting quantitative characteristics of the drivability represented by the intelligent vehicle, and establishing a training set database; combining a Delphi method and a subjective evaluationengineer knowledge base to construct a subjective and objective correlation analysis and drivability automatic evaluation standard evaluation system; and by means of a machine learning C4.5 algorithm, performing analysis of subjective and objective correlation of the intelligent vehicle and automatic evaluation of drivability. The method can provide an evaluation basis for the feasibility of a decision planning algorithm and a tracking control algorithm of the intelligent vehicle.

Description

technical field [0001] The invention belongs to the technical field of performance evaluation and analysis of intelligent vehicles, and in particular relates to an automatic evaluation method for driving performance of intelligent vehicles by fusing multi-source vehicle sensor information. Background technique [0002] Intelligent vehicles integrate environmental perception, decision-making and planning, and trajectory tracking modules, and use radar, cameras, and planning control algorithms to simulate the driver's driving behavior to ensure the safety, economy, and comfort of the vehicle-road-human system. and the focus and hotspots of the development of the field of robotics. When evaluating the decision-making planning algorithm and tracking control algorithm, only the safety and economy are the goals, and the vehicle execution ability and the subjective feelings of the drivers and passengers are not considered, which makes the ride comfort of the vehicle poor. Therefor...

AI Technical Summary

Problems solved by technology

The current patents and research on drivability evaluation focus on subjective evaluation and traditional vehicles, but the subjective evaluation of drivability is costly, has a large environmental impact, is difficult to repeat, and is difficult to provide for the optimization of intelligent vehicle trajectory planning and tracking control algorithms. objective reference value

Images