As vehicles evolve from driver assistance toward full autonomy, the boundary between human and machine control becomes increasingly critical. In Level 2 and Level 3 autonomous systems, maintaining safe and intuitive transitions between manual and automated driving presents both a technical and human factors challenge. This presentation explores a high-quality agent framework designed to mediate shared autonomy—intelligently delegating control between human drivers and autonomous systems based on context, capability, and risk. Drawing from simulation studies and on-road testing, we illustrate how this collaborative approach enhances trust, safety, and user experience. The session will highlight decision arbitration techniques, real-time driver state monitoring, and adaptive control strategies that ensure seamless handovers and effective co-driving between human and machine.

In this session, you will learn more about:

• How high-quality control agents can dynamically manage the delegation between manual and autonomous driving in L2/L3 systems
• The importance of human-machine interface (HMI) design and driver state assessment in enabling safe and intuitive shared control
• Strategies to improve system transparency and trustworthiness through context-aware handover policies