Super-Ellipse Control Barrier Functions for Enhanced Obstacle Avoidance in Complex Multi-Agent Environments
Wenbin Liu, Shucheng Jia, Mikhail M. Svinin · 2025
Autonomous mobile robots have rapidly evolved in recent years, serving critical roles in fields such as logistics, medical services, and urban construction. While Control Barrier Functions (CBF) have demonstrated effectiveness in ensuring real-time safety and collision avoidance, they often exhibit limitations in irregular or dynamic environments, leading to overly conservative maneuvers. In this paper, we propose a novel Super-Ellipse Control Barrier Function (SE-CBF) framework that substantially improves upon conventional CBF by dynamically modulating safety distances around obstacles. This adaptive approach not only accommodates non-uniform obstacle shapes but also optimizes path utilization in multi-agent scenarios. The super-ellipse design allows robots to flexibly enlarge or shrink their safety zones along different axes, minimizing path redundancy and improving navigation efficiency. Extensive simulations verify the robustness and effectiveness of our method. Moreover, real-world experiments with simplified multi-obstacle intersections confirm that SE-CBF outperforms traditional CBF in both safety assurance and operational stability. Our findings demonstrate that SE-CBF can serve as a pivotal step toward safer and more efficient autonomous navigation in complex, high-density domains.