Distributionally Robust Acceleration Control Barrier Filter for Efficient UAV Obstacle Avoidance

TL;DR

The paper tackles the challenge of rapid, safe dynamic obstacle avoidance for UAVs under perception latency and limited onboard compute. It introduces DR-ACBF, an acceleration-space safety filter that enforces linear half-space constraints, augmented with Cantelli tightening and a DR-CVaR early warning, and realized via a constant-time Gauss-Southwell projection instead of solving a QP. The approach achieves comparable safety performance to traditional methods while significantly reducing computation time, as shown in simulations, and demonstrates feasibility on a Crazyflie in hardware experiments. These contributions offer a practical, provably safe DOA framework suitable for high-rate, resource-constrained UAV platforms.

Abstract

Dynamic obstacle avoidance (DOA) for unmanned aerial vehicles (UAVs) requires fast reaction under limited onboard resources. We introduce the distributionally robust acceleration control barrier function (DR-ACBF) as an efficient collision avoidance method maintaining safety regions. The method constructs a second-order control barrier function as linear half-space constraints on commanded acceleration. Latency, actuator limits, and obstacle accelerations are handled through an effective clearance that considers dynamics and delay. Uncertainty is mitigated using Cantelli tightening with per-obstacle risk. A DR-conditional value at risk (DR-CVaR)based early trigger expands margins near violations to improve DOA. Real-time execution is ensured via constant-time Gauss-Southwell projections. Simulation studies achieve similar avoidance performance at substantially lower computational effort than state-of-the-art baseline approaches. Experiments with Crazyflie drones demonstrate the feasibility of our approach.

Figures (3)

• Figure 1: Illustration of the proposed DR-ACBF framework. The UAV avoids dynamic obstacles (top) by constraining its acceleration within distributionally robust half-spaces marked in grey. The UAV view (bottom) depicts the projected half-space plane towards the obstacles. The effective clearance $R_{\mathrm{eff}}$ accounts for latency and actuation limits, while the obstacles' uncertainty ellipsoids $\sigma_{dH,i}$ and risk levels $\alpha_i$ tighten the constraints under sensing uncertainty. The safe set of action $C$ ensures that the UAV's safe zone is not violated.
• Figure 2: System block diagram. The process begins with sensing the UAV and obstacle states. UAV control generates a nominal acceleration command $a_n$, which is corrected by DR-ACBF avoidance. The DR-CVaR layer provides an early warning, the acceleration estimation, and risk allocation to compute the safety constraints. Finally, ACBF and projection replace $a_n$ with the safe acceleration command $a_e$ to avoid obstacles.
• Figure 3: Experimental lab setup. (Left) The green Crazyflie UAV must avoid two dynamic obstacles (marked in red): an approaching UAV and a thrown object. (Right) The motion-capturing system provides position data of the UAV and the obstacles. A video of our experiments is available for review.