Hierarchical Switching CBFs: A Safety Framework for Complex Lane-Changing Scenarios

Ensuring the safety and efficiency of autonomous vehicles in increasingly complex, dynamic, and structured road environments remains a key challenge. While traditional optimization-based approaches can provide safety guarantees, they often struggle to meet real-time requirements due to high computational complexity. Concurrently, although Control Barrier Functions (CBFs) can ensure instantaneous safety with minimal intervention, their inherent locality makes it difficult to consider global task objectives, potentially leading to mission failure in complex scenarios like lane-change obstacle avoidance. To address this trade-off between safety and mission completion, this paper proposes a hierarchical switching CBF safety framework. The core of this framework is to decompose complex lane-change tasks into multiple logical phases and to activate specialized, pre-designed CBF constraint sets via a top-level logic controller. Finally, we demonstrate the feasibility and safety of the proposed framework in urban driving scenarios through MATLAB simulations.