TY - JOUR
T1 - A human–machine shared dual fuzzy authority allocation control strategy for automatic driving vehicle considering driver intention judgement
AU - Wang, Weida
AU - Zhang, Yunpu
AU - Yang, Chao
AU - Zhang, Yuhang
AU - Gao, Yipeng
AU - Ma, Taiheng
AU - Qie, Tianqi
N1 - Publisher Copyright:
© 2025 Elsevier Ltd
PY - 2025/5/15
Y1 - 2025/5/15
N2 - Human–machine co-driving is a driving process of human and machine complete driving tasks together in advanced driving assistance system, achieving a smooth transition from automatic driving to manual driving. In this process, both human driver and automatic system remain in the control loop according to their respective control intentions. Human–machine shared control achieves the dynamic allocation of driving authorities. Therefore, the intention consistency and driving authority allocation strategy of human-machine will affect vehicle driving safety. How to allocate control authority in consideration of human–machine intentions is a key challenge. To address this issue, this paper proposes a shared control strategy based on dual fuzzy authority allocation considering driver intention judgement. First, as the basis of driving authority allocation, the driver takeover request recognition scheme based on the torque threshold considering real driver characteristics is proposed. The threshold is determined by data collected on a steer-by-wire test bench and processed through K-means clustering algorithm. Then, based on human–machine relationships and quantitative evaluation of driving characteristics, a dual fuzzy authority allocation controller with different emergency and auxiliary correction rules is designed to allocate control authorities. Finally, the proposed strategy is verified across diverse scenarios by driver-in-the-loop tests. Results indicate the proposed strategy reduces emergency obstacle avoidance time by 35% in conflicting scenarios, while reducing lateral offset and yaw errors by 53.89% and 11.11% in cooperative scenarios, compared to fixed-weight authority allocation strategy.
AB - Human–machine co-driving is a driving process of human and machine complete driving tasks together in advanced driving assistance system, achieving a smooth transition from automatic driving to manual driving. In this process, both human driver and automatic system remain in the control loop according to their respective control intentions. Human–machine shared control achieves the dynamic allocation of driving authorities. Therefore, the intention consistency and driving authority allocation strategy of human-machine will affect vehicle driving safety. How to allocate control authority in consideration of human–machine intentions is a key challenge. To address this issue, this paper proposes a shared control strategy based on dual fuzzy authority allocation considering driver intention judgement. First, as the basis of driving authority allocation, the driver takeover request recognition scheme based on the torque threshold considering real driver characteristics is proposed. The threshold is determined by data collected on a steer-by-wire test bench and processed through K-means clustering algorithm. Then, based on human–machine relationships and quantitative evaluation of driving characteristics, a dual fuzzy authority allocation controller with different emergency and auxiliary correction rules is designed to allocate control authorities. Finally, the proposed strategy is verified across diverse scenarios by driver-in-the-loop tests. Results indicate the proposed strategy reduces emergency obstacle avoidance time by 35% in conflicting scenarios, while reducing lateral offset and yaw errors by 53.89% and 11.11% in cooperative scenarios, compared to fixed-weight authority allocation strategy.
KW - Automatic driving vehicle
KW - Driver intention judgement
KW - Dual fuzzy authority allocation control
KW - Human–machine shared control
UR - http://www.scopus.com/inward/record.url?scp=85218865652&partnerID=8YFLogxK
U2 - 10.1016/j.eswa.2025.126971
DO - 10.1016/j.eswa.2025.126971
M3 - Article
AN - SCOPUS:85218865652
SN - 0957-4174
VL - 274
JO - Expert Systems with Applications
JF - Expert Systems with Applications
M1 - 126971
ER -