Driver’s Intention Identification With the Involvement of Emotional Factors in Two-Lane Roads

Xiaoyuan Wang,Shanliang Liu,Yongqing Guo,Quan Yuan,Chenglin Bai,Junyan Han

doi:10.1109/tits.2020.2995837

Abstract

Driver’s emotion is a psychological reaction to environmental stimulus. Driver intention is an internal state of mind, which directs the actions in the next moment during driving. Emotions usually have a strong influence on behavioral intentions. Therefore, emotion is an important factor that should be considered, to accurately identify driver’s intention. This study used the support vector machine (SVM) theory to develop a driver intention recognition model, with the involvement of driver’s emotions. Various materials including visual materials, auditory materials, and olfactory materials, were used to induce driver’s emotions. Real driving, virtual driving and computer simulation experiments were conducted to collect human-vehicle-environment dynamic data in two-lane roads. The results present that the proposed model can achieve high accuracy and reliability in recognizing driver’s intentions. Our findings of this study can be used to develop the personalized driving warning system and intelligent human-machine interaction in vehicles. This study would be of great theoretical significance for improving road traffic safety.

Highlights

C URRENT driver-assistance systems evaluate safety situations and potential hazards, mainly relying on traffic information inputs from Lidar or visual sensors
Regarding lane-changing intention recognition shown in Figure13, the model can achieve an accuracy of above 85% for the states of fear, anxiety, contempt and anger, over 80% for the states of helplessness and surprise, as well as 75% or higher for the states of relief and pleasure
The results indicate that the proposed identification model can obtain high accuracy in recognizing intentions of driving speed and lane change

Summary

Introduction

C URRENT driver-assistance systems evaluate safety situations and potential hazards, mainly relying on traffic information inputs from Lidar or visual sensors. Without considering the impacts on driving behavior, driver-assistance systems may provide false or unnecessary alerts [2]. This issue raises driver’s annoyance and reduces trust, as a result, increasing the chance of traffic accident. Berndt et al [3] used the Hidden Markov Model (HMM) to identify driver’s intentions of turning and going-straight, based on driving data such as acceleration, pedal position, brake pressure, and steering wheel angle. Ürün et al [5] used the artificial neural network model and support vector machine to predict driver’s behavior, based on different combinations of driving and road data including road curvature, lane position, steering wheel angle, lateral acceleration, and collision time. Melnicuk et al [11] adopted repeated measure design and

Methods

Results

Discussion

Conclusion