Pavement adhesion coefficient estimation method based on Bayes classifier

Abstract

The invention discloses a pavement adhesion coefficient estimation method based on a Bayes classifier, and the method is based on the theory of the Bayes classifier and comprises the steps of determining the attribution of a road model through a data point set of the traction force of a certain side and the slip rate of an automobile during the driving of the automobile, dividing the collected overall data point set into a plurality of data subsets according to the traction force range, analyzing the newly collected data points, obtaining the prior probability of the data points, calculating the posterior probability of the data points, determining the attribution of the road model, and accurately calculating the adhesion coefficient of the road model according to the new road model. Through the above mode, by learning the prior probability of the Bayes subset, when a real-time data point arrives, the posterior probability of each traction subset interval is calculated, the road surface switching time can be calculated through the method, and the point set affiliation result after road surface switching can be obtained, so that the affiliation of the road model is determined; and the adhesion coefficient of the pavement model is obtained.

Description

technical field [0001] The invention relates to the field of calculation of road surface adhesion coefficient, in particular to a method for estimating road surface adhesion coefficient based on Bayes classifier. Background technique [0002] At present, the number of domestic automobiles is increasing rapidly, and safety performance has become the most important consideration for automobiles. In the increasingly mature field of autonomous driving, the identification of road adhesion coefficient is becoming more and more important. The braking distance of the car on different adhesion roads is different. The braking distance is longer on the road with a small adhesion coefficient, and the stability of the car is less controllable. Therefore, when the road adhesion coefficient is switched, it is faster to identify the change of the adhesion coefficient Appears to be extremely important. [0003] In the existing sensor-based road surface classification recognition and switch...

AI Technical Summary

Problems solved by technology

[0006] In order to overcome the problem that the existing road surface switching recognition system has poor real-time recognition and the algorithm convergence speed is greatly affected by the road surface conditions, resulting in inaccurate calculation data, the present invention provides a road surface adhesion coefficient estimation method based on Bayes classifier, The prior probability can be obtained by learning the road surface with different adhesion coefficients, and then add the idea of ​​Bayes classifier to classify the subsequent data sets, and calculate the posterior probability through the prior probability to judge the data sets that have come Which type of road surface model belongs to, and then the adhesion coefficient of the road surface model can be accurately calculated through the Kalman filter algorithm, which can realize the switching recognition of the road surface during the driving process of the car and the calculation of the adhesion coefficient of the switched road surface model

Images