Adaptive authority allocation for shared steering control considering social behaviours: a hybrid fuzzy approach with game-theoretical vehicle interaction model

Abstract

In a typical traffic scenario, automated vehicles are required to interact with surrounding traffic participants, e.g. human-driven vehicles, pedestrians, etc. In this paper, a novel adaptive authority allocation strategy considering the social behaviours of surrounding vehicles is proposed for the shared steering control (SSC) of automated vehicles. First, a Koopman-based potential-field-driven distributed model predictive control (K-PF-DMPC) method is proposed for the modelling of vehicle interaction to describe surrounding vehicle's social behaviour. This method effectively deals with the nonlinearity embedded in the non-cooperative game-based vehicle interaction model, so that the analytical form of Nash equilibrium can be derived for a fast online solution. Then, the SSC system of the automated vehicle is implemented by a weighted summation method. To balance the driving performance and intervention degree while considering the surrounding vehicles' social behaviours, a hybrid fuzzy strategy (HFS) is proposed for allocating the control authority between the driver and automation. The value of authority allocation coefficient is calculated by fusing the outputs from two shared fuzzy controllers, i.e. the enhanced and weakened shared fuzzy controllers, in accordance to the belief of the surrounding vehicle's type. Several numerical simulations and driver-in-the-loop simulator experiments are conducted for validation. Results show that the proposed strategy can provide appropriate steering intervention to the driver owing to the consideration of the surrounding vehicle's social behaviour and obtains the best driver-automation collaboration performance compared with the comparative groups.

Keywords:

• Shared steering control
• adaptive authority allocation
• social behaviours
• Koopman-based distributed model predictive control
• hybrid fuzzy strategy

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the National Natural Science Foundation of China (NSFC) under grant number 51975194, the Natural Science Foundation of Hunan Province under grant number 2021JJ30121, and the State Key Laboratory of Automotive Safety and Energy under Project No. KFZ2203.