Human-Machine Shared Driving Control for Semi-Autonomous Vehicles Using Level of Cooperativeness

Area of Science:

• Automotive Engineering
• Control Systems
• Human-Machine Interaction

Background:

• Semi-autonomous vehicles require effective human-automation collaboration for safe and efficient operation, particularly in lane-keeping tasks.
• Existing shared control systems often struggle to adapt dynamically to varying driver states and vehicle conditions.
• Understanding human-machine interaction principles is crucial for designing cooperative driving systems.

Purpose of the Study:

• To propose and validate a new haptic shared control concept for lane keeping in semi-autonomous vehicles.
• To introduce a metric for human-machine cooperative status to dynamically adjust automation assistance.
• To manage driver workload and performance characteristics for optimized control authority.

Main Methods:

• Development of a human-machine cooperative status metric and driver workload assessment.
• Design of a time-varying assistance factor modulating torque based on driver performance.
• Implementation of an integrated driver-in-the-loop vehicle model including yaw-slip, steering, and driver dynamics.
• Application of a novel ℓ∞ linear parameter varying control technique to handle time-varying parameters.
• Validation using Lyapunov stability theory and high-fidelity simulations.

Main Results:

• The proposed haptic shared control method effectively manages driver-automation conflicts during lane keeping.
• The system dynamically adjusts haptic assistance based on real-time driver cooperation and workload.
• High-fidelity simulations demonstrate the robustness and effectiveness of the control strategy across various driving scenarios.
• The ℓ∞ linear parameter varying control technique successfully addresses the time-varying nature of the system.

Conclusions:

• The novel haptic shared control concept offers a significant advancement in cooperative driving for semi-autonomous vehicles.
• The adaptive approach enhances safety and performance by optimizing the balance between human control and automation.
• This research provides a robust framework for future development of intelligent driver-assistance systems.