Driving style identification method for optimizing BP neural network based on improved genetic algorithm

Abstract

The invention discloses a driving style identification method for optimizing a BP neural network based on an improved genetic algorithm, and the method comprises the steps: collecting the driving dataof a tested driver through an experiment and simulation, screening driving data of each driver, removing unreasonable driving data, and extracting a feature parameter which can reflect the driving style of each driver; clustering the driving data on the basis of a Kmean clustering algorithm, and defining the driving types of the drivers as a aggressive type, a general type and a prudent type according to a clustering result; reading a driving type and a characteristic parameter value of the driver obtained by clustering in S3 to construct a topological structure of a BP neural network, optimizing a weight and a threshold value of the BP neural network by using an improved genetic algorithm, and assigning the optimized weight and threshold value of the BP neural network to the BP neural network; finally, training and testing the optimized BP neural network and recognizing the driving style.

Description

technical field [0001] The invention belongs to the technical field of driving style recognition, and in particular relates to a driving style recognition method based on an improved genetic algorithm to optimize a BP neural network. Background technique [0002] Advanced Driver Assistance System (ADAS) refers to a system that relies on on-board sensing systems to perceive the environment and assist the driver in driving operations. The driving assistance system is beneficial to reduce the occurrence of traffic accidents, improve driving safety, and improve traffic capacity. However, the current design mode of the assisted driving system is single, which cannot meet the individual differences of different drivers, and reduces the driver's acceptance and satisfaction with the driving assisted system. Driver styles are generally divided into three categories: aggressive, general, and cautious. Drivers with different driving styles will show different reactions to the same dri...

AI Technical Summary

Problems solved by technology

Among them, the BP neural network has the advantages of strong nonlinear mapping ability, high self-learning and self-adaptive ability, etc., but the core idea based on the steepest descent method leads to the inevitable existence of BP network that is easy to fall into local minimum, low generalization ability, and slow convergence speed. Slower and other disadvantages

[0004] Genetic Algorithm (GA) is an intelligent algorithm for optimization. It can optimize the weight and threshold of BP neural network through genetic algorithm to improve the recognition rate, but the standard genetic algorithm is easy to fall into local extremum and converge. slow

Images