Study on Driver Model Parameters Distribution for Fatigue Driving Levels Based on Quantum Genetic Algorithm

ABSTRACT

According to the assumption that fatigue study cannot reveal fatigue mechanism and nonlinear influence factors of vehicle driving closed-loop system defects, this paper proposes a driver model inversion method for studying the driver's fatigue diagnosis. Furthermore, the new method is divided into two steps: 1. By using the forecast of neural network model to build the driver-vehicle-road closed-loop model, which is adapted to the complex road conditions. Besides, and the model was used to study the changes in the closed-loop car system parameter in which the driver is in a state of fatigue. 2. By defining specific movement track through the degree of approximation of theoretical data and taking test data as the objective function, the driver parameter inverse problem was broken into multiple target optimization problems. A method of real-coded chaotic mutation of quantum genetic algorithm (GA) optimization is used to find the global optimal solution. The driving simulation test results show that under the condition of complex road conditions, the proposed algorithm in actual driving parameter inversion of the alignment is superior to the traditional genetic algorithm (GA) and the traditional quantum genetic algorithm (QGA). Finally, the relationship between pilot model parameters and fatigue factors is established.

Keywords: