Human–Machine Adaptive Shared Control for Safe Driving Under Automation Degradation

In this article, a human–machine adaptive shared control method is proposed for automated vehicles (AVs) under automation performance degradation. First, a novel risk assessment module is proposed to monitor driving behavior and evaluate automation performance degradation for AVs. Then, an adaptive control authority allocation module is developed. In the event of any performance degradation, the control authority allocated to the automation system is decreased based on the assessed risk. Consequently, the control authority allocated to a human driver is adaptively increased and thus requires more driver engagement in the control loop to compensate for the automation degradation and ensure the vehicle’s safety. Experimental validation is conducted under different driving scenarios. The test results show that the approach can effectively compensate for vehicle automation performance degradation through human–machine adaptive shared control, ensuring the safety of automated driving.

In recent years, research institutes, automotive manufacturers, and IT companies have made a great effort to develop autonomous driving technology, which has great potential to improve the safety and efficiency of transportation systems [1]. Although fully autonomous driving is the goal, many technical problems remain to be solved, and highly automated vehicles are expected to play a significant role in the interim [2]. Highly automated driving, which retains a human driver in the control loop, presents an exciting new development in vehicle technology [3], [4]. The approach allows a human driver and an automation system to share control authority and cooperatively operate a vehicle [5], [6]. This poses a new challenge, namely, how to ensure safe and smooth control allocations and transitions between humans and automation systems.