Modeling of Driver's Collision Avoidance Maneuver Based on Controller Switching Model

Abstract

Abstract-This paper presents a modeling strategy of human driving behavior based on the controller switching model focusing on the driver's collision avoidance maneuver. The driving data are collected by using the three-dimensional driving simulator based on CAVE, which provides stereoscopic immersive virtual environment. In our modeling, the control scenario of the human driver, that is, the mapping from the driver's sensory information to the operation of the driver such as acceleration, braking and steering, is expressed by Piecewise Polynomial (PWP) model. Since the PWP model includes both continuous behaviors given by polynomials and discrete logical conditions, it can be regarded as a class of Hybrid Dynamical System (HDS). The identification problem for the PWP model is formulated as the Mixed Integer Linear Programming (MILP) by transforming the switching conditions into binary variables. From the obtained results, it is found that the driver appropriately switches the 'control law' according to the sensory information. Also, the driving characteristics of the beginner driver and the expert driver are compared and discussed. These results enable us to capture not only the physical meaning of the driving skill, but also the decision-making aspect (switching conditions) in the driver's collision avoidance maneuver.

Modeling and Recognition of Driving Behavior Based on Stochastic Switched ARX Model

“…The measured profiles of the sensory information and steering operation for the seven drivers are shown in Figs.6,7,8,and 9. The Ai, Bi and Ci in each figure denote the corresponding driver's trial index.…”

“…Due to the abovementioned circumstances, driving behavior modeling has attracted much attention from several researchers [1][2] [3][4] [9] [10]. In order to model driving behavior, the conventional techniques such as the nonlinear regression models, neural networks and fuzzy systems have been used [5] [6].…”

“…In [9] and [10], we applied the HDS with deterministic mode change to the modeling of driving behavior. Although this study could capture the motion and decision making aspects in human behavior, it is not considered suitable for real-time behavior recognition due to its high computational cost.…”

“…Although this study could capture the motion and decision making aspects in human behavior, it is not considered suitable for real-time behavior recognition due to its high computational cost. This is because mixed integer programming and data clustering technique were used for the parameter estimation in [9] and [10], respectively.…”

Study of Correlation between Driver Emergency Measures and Pedestrian Injury Based on Combined Driving Simulator and Computer Simulation

“…Xuemei Chen's study, which is about drivers' braking behaviors under emergency, showed that the control of brake pedal under emergency can be divided into four stages [13]. Similarly, Kim et al studied different emergency measures among different driving experiences, which discussed the factors of steering wheel angle and brake pedal pressure, and concluded that steering is a more advanced emergency decision-making method than braking [14]. However, if we have not the records for driving, it remains hard to take drivers' emergency reaction into consideration in existing accident analysis.…”