Development of Autonomous Driving Techniques for Prevention of Multiple Rear-End Collision: Vehicle Platoon Safety by Intent Inference of Driver's Deceleration Behaviour

2021 Fiscal Year Final Research Report

• Project/Area Number: 19K04926
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 25020:Safety engineering-related
• Research Institution: Nippon Institute of Technology
• Principal Investigator: SUZUKI HIRONORI 日本工業大学, 先進工学部, 教授 (20426258)
• Project Period (FY): 2019-04-01 – 2022-03-31
• Keywords: 意図推定 / 動的推定 / アンセンティッドカルマンフィルタ / ショックウェーブ / 車群交通流制御 / 運転支援 / 自動運転

Outline of Final Research Achievements

This research aims to propose a new vehicle platoon control technology with the prevention of multiple rear-end collisions and to verify its effectiveness. The objectives were to establish a technique for predicting the deceleration intention of the vehicle ahead, to propose a driving assistance or autonomous driving technique to encourage early deceleration of the vehicle behind, and to propose a vehicle platoon control technique to prevent multiple rear-end collisions. Mathematical modelling and numerical simulation showed that the fusion of an unscented Kalman filter and a neural network model successfully predicted deceleration intentions several seconds ahead with high accuracy. Furthermore, it is confirmed that the utilizing the visualized shockwave contributed to suppression of shockwave propagation and prevention of multiple rear-endo collisions with simple control techniques.

Free Research Field

交通予防安全工学

Academic Significance and Societal Importance of the Research Achievements

高速道路など高密度・高速度で走行する車群内での多重追突事故の被害は著しいことから，追突事故の抑制は社会安全上の喫緊の課題である。この研究の遂行により，学術的には，数理モデリングの融合により車群を安全化・安定化する技術の開発に成功し，この成果は，死亡事故に至るようなリスクの高い交通事故の低減に寄与する可能性があると言う点において，その社会的意義も大きい。