Decision-Making Research for Intelligent Connected Vehicles Considering Adhesion Condition Research Institute of Highway, Ministry of Transport

Format:

Book

Conference/Event

Author/Creator:

Wang, Hong, author.

Contributor:

Hou, De-Zao

Conference Name:

2025 5th International Conference on Smart City Engineering and Public Transportation (SCEPT2025) (2025-03-28 : Beijing, China)

Language:

English

Physical Description:

1 online resource cm

Place of Publication:

Warrendale, PA SAE International 2025

Summary:

The adhesion condition of the road surface is an important factor in the driving decision-making, and the lower the adhesion coefficient of the road, the greater the risk of safety. In order to study the development and progress in the research of the substances, a comparative analysis of Chinese and foreign references was carried out. The sensitive factors to the adhesion coefficient and influence of adhesion condition on driving were summarized. Then two main strategies to avoid a collision were presented, including longitudinal braking and lateral lane change. A detailed description of three methods used in automotive decision-making processes was offered, including rule-based method, supervised learning method, and reinforcement learning method, each characterized with certain attributes. Topics in the field of driving decision-making considering adhesion condition for intelligent connected vehicles were pointed out and future-oriented research formulations were provided. These results indicate that (1) Factors about roads are lacking in driving behaviour decision-making research. Thus, the studies on driving decision-making under different road conditions need to be carried out to improve the adaptability of driving decision-making systems in complex road environment. (2) Studies on cooperative driving decision-making in intelligent connected vehicles are important to enhance the efficiency in dealing with various complicated and dangerous traffic environments. There is a need for extended work on cooperative driving strategies. (3) There are also some issues in reinforcing learning with the driving decision-making, such as difficulty of realization. Accordingly, the introduction of human feedback-driven reinforcement learning is favored in the study about driving decision-making. This method is believed to improve the decision-making performance of systems and speed up the transition of research to the real-world interception

Notes:

Vendor supplied data

Publisher Number:

2025-99-0028

Access Restriction:

Restricted for use by site license