Distributed Control Barrier Functions for Safe Multi-Vehicle Navigation in Heterogeneous USV Fleets

TL;DR

Distributed control barrier functions are used to enforce safety by ensuring forward-invariance of the safe set $\\mathcal{C}$ in a distributed per-vehicle framework, even when other agents may act adversarially. A quadratic program (QP) based control filter computes each vehicle's input by respecting pairwise CBF constraints $L_g h_{pair} u_i + \zeta^{min} + \alpha(h_{pair}) \ge 0$ and by intersecting across neighbors to form $\\mathcal{C}_{\\cap}$ and $K_{cbf_i}$. The approach is validated through both large-scale simulations and real-world joust experiments with four heterogeneous platforms including a crewed vessel, showing that CBFs eliminate collisions while COLREGS improves efficiency, and that combining both yields the best safety-performance trade-off. The findings support safe, explainable deployment of heterogeneous USV fleets in mixed autonomous-human environments without requiring sharing intent or full trajectory information.

Abstract

Collision avoidance in heterogeneous fleets of uncrewed vessels is challenging because the decision-making processes and controllers often differ between platforms, and it is further complicated by the limitations on sharing trajectories and control values in real-time. This paper presents a pragmatic approach that addresses these issues by adding a control filter on each autonomous vehicle that assumes worst-case behavior from other contacts, including crewed vessels. This distributed safety control filter is developed using control barrier function (CBF) theory and the application is clearly described to ensure explainability of these safety-critical methods. This work compares the worst-case CBF approach with a Collision Regulations (COLREGS) behavior-based approach in simulated encounters. Real-world experiments with three different uncrewed vessels and a human operated vessel were performed to confirm the approach is effective across a range of platforms and is robust to uncooperative behavior from human operators. Results show that combining both CBF methods and COLREGS behaviors achieves the best safety and efficiency.

Figures (10)

• Figure 1: Fleet of USVs using the distributed control barrier function (CBF) method described in this paper while performing a search mission on the Charles River. A graphical depiction of the barriers are shown by the red circles.
• Figure 2: State definition for the $i^{th}$ or ego vehicle
• Figure 3: Example of mapping of constraints in state space to control input space and the process of control filtering. Left: a vehicle must remain within the safe set $\mathcal{C}$ marked as the green areas. This state constraint is translated into constraints in the control space, rendering some combinations of the rudder and thrust unsafe. The nominal control is unsafe and is filtered to the closest safe control input.
• Figure 4: The combined safe set $\mathcal{C}_\cap$ (green) is the intersection of all the pairwise safe sets $\mathcal{C}$.
• Figure 5: Overview of the joust mission used for structured evaluation of safety. Each vehicle starts at one side of a 64 meter circle and at the same time they all begin to travel to their own goal point on the other side of the circle, maneuvering around the other vehicles in their way. Once they have reached the other side, each vehicle executes a 360 degree turn to prepare for the next cycle.