A workload adaptive haptic shared control scheme for semi-autonomous driving.

Area of Science:

• Human-computer interaction
• Autonomous systems
• Automotive engineering

Background:

• Haptic shared control allocates control between humans and autonomous agents in semi-autonomous driving.
• Current systems often neglect the human operator's state, limiting effectiveness.

Purpose of the Study:

• To develop and evaluate a novel haptic shared control scheme that dynamically adapts to human operator workload, visual attention (eyes on road), and steering input torque.
• To investigate the impact of this adaptive scheme on human-machine interaction and driving performance.

Main Methods:

• Human-in-the-loop experiments were conducted with 24 participants controlling a simulated High Mobility Multipurpose Wheeled Vehicle (HMMWV).
• Participants shared control with an autonomous navigation module while performing a concurrent target detection task.
• The control system was tested in both adaptive and non-adaptive configurations based on real-time human factors.

Main Results:

• The adaptive haptic control scheme significantly reduced operator workload compared to the non-adaptive system.
• Participants reported higher trust in the adaptive autonomy.
• Driving task performance improved, and required control effort was reduced with the adaptive scheme.

Conclusions:

• Adaptive haptic shared control that considers real-time human factors offers substantial benefits over non-adaptive approaches.
• This adaptive strategy enhances safety, efficiency, and user acceptance in semi-autonomous driving systems.
• Future research should explore further personalization and integration of adaptive shared control in diverse driving scenarios.