Patent · US Active

Methods for cooperative decision-making on lane-changing behaviors of autonomous vehicles based on bayesian game

US12134409B1 · kind B1 · utility

0Cited by
0References
8Claims
0Family size

Assignee

Inventors

Key dates

Filing dateJun 25, 2024
Grant dateNov 5, 2024
Priority date
Expiry dateJun 25, 2044

Classification

  • Technology area (CPC Y)Emerging Cross-Sectional Technologies
  • CPC primaryY02T10/40
  • WIPO fieldTransport
  • WIPO sectorMechanical engineering

Abstract

Disclosed is a method for cooperative decision-making on lane-changing behavior of an autonomous vehicle based on Bayesian game. On one hand, intelligent networked road perception and big data analysis are utilized to infer statistical characteristics of driving styles of a side vehicle under different time periods and traffic flow states, serving as prior predictions of driving styles of the side vehicle. On the other hand, the dynamic interaction behaviors during lane-changing processes of a specified vehicle and the side vehicle are continuously observed and posterior corrections are made to the driving styles of the side vehicle. When the specified vehicle generates a lane-changing willingness, probabilities of driving styles are iteratively predicted using Bayesian game principles. Comprehensive consideration of style and willingness probabilities yields a lane-changing probability of the specified vehicle. Once the lane-changing probability exceeds a threshold, a lane-changing activation instruction is issued.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.