Fair and Safe: A Real-Time Hierarchical Control Framework for Intersections

TL;DR

This paper tackles the challenge of fair and safe real-time intersection management for connected and automated vehicles by proposing a two-layer hierarchical framework. The top layer employs an inequity-aversion utility to allocate control authority to a single vehicle, balancing waiting time, control history, and urgency to achieve near-perfect fairness. The bottom layer executes a precomputed DDP trajectory with a curvature-aware LQR/PD tracker and a flexible high-order CBF safety filter to guarantee collision-free operation under dynamic obstacles. Across unsignalized and signalized intersections, simulations show substantial gains in fairness and efficiency with zero collisions, while maintaining real-time feasibility (frequencies well above 100 Hz in most cases). The approach offers a scalable, modular solution that integrates offline optimal planning with online safety guarantees, supporting future autonomous traffic systems with equitable access and strong safety performance.

Abstract

Ensuring fairness in the coordination of connected and automated vehicles at intersections is essential for equitable access, social acceptance, and long-term system efficiency, yet it remains underexplored in safety-critical, real-time traffic control. This paper proposes a fairness-aware hierarchical control framework that explicitly integrates inequity aversion into intersection management. At the top layer, a centralized allocation module assigns control authority (i.e., selects a single vehicle to execute its trajectory) by maximizing a utility that accounts for waiting time, urgency, control history, and velocity deviation. At the bottom layer, the authorized vehicle executes a precomputed trajectory using a Linear Quadratic Regulator (LQR) and applies a high-order Control Barrier Function (HOCBF)-based safety filter for real-time collision avoidance. Simulation results across varying traffic demands and demand distributions demonstrate that the proposed framework achieves near-perfect fairness, eliminates collisions, reduces average delay, and maintains real-time feasibility. These results highlight that fairness can be systematically incorporated without sacrificing safety or performance, enabling scalable and equitable coordination for future autonomous traffic systems.

Figures (5)

• Figure 1: Overview of the proposed hierarchical control framework. The system first generates reference trajectories using offline DDP-based optimal motion planner. Then, a fairness-oriented control authority allocation module—driven by an inequity aversion utility—rapidly determines the authorized vehicle, enabling hybrid centralized planning and decentralized execution. Finally, the real-time trajectory tracking controller executes the planned trajectory while ensuring safety through LQR/PD control, repulsive fields, and a flexible high-order CBF-based safety filter.
• Figure 2: Intersection simulations under medium-demand scenarios and highly unbalanced conditions, with the proposed fairness-aware framework.
• Figure 3: Jain's fairness index across traffic scenarios
• Figure 4: Gini coefficient across traffic scenarios
• Figure 5: Real-time control frequency distribution