FastBridge: Closing the Model-Based Realization Gap in Safety Filters on 3D Gaussian Splatting for Fast Quadrotor Flight

📅 2026-07-01
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
This work addresses the challenge of high-speed quadrotor flight under limited onboard computational resources, where simultaneously satisfying actuator constraints and ensuring real-time safe obstacle avoidance remains difficult. The authors propose a nonlinear safety filter that incorporates full quadrotor dynamics, leveraging 3D Gaussian Splatting to construct a continuous geometric representation of the environment. For the first time, they introduce a high-relative-degree collision cone exponential control barrier function combined with a backup control barrier function, integrated with a forward-simulation-based backup strategy to guarantee feasibility of the input-constrained quadratic program (QP). This approach bridges the model-implementation gap between perception and control, enabling stable and real-time navigation in complex environments. Evaluated in both simulation and physical hardware, the method reduces trajectory jitter by 47% and achieves a 2.25× speedup compared to existing approaches.
📝 Abstract
Fast quadrotor flight requires safe obstacle avoidance under tight onboard compute limits. While 3D Gaussian Splatting (3DGS) provides a continuous, geometry-aware scene representation for perception-driven navigation, existing 3DGS safety filters use reduced-order models such as single- and double-integrators that ignore actuator limits and assume commanded accelerations are realized instantaneously. Building on an analytic collision cone barrier for 3DGS, we introduce a nonlinear, actuator-aware safety filter enforced through the full quadrotor dynamics. We derive a high-relative-degree collision cone exponential CBF and a backup CBF that preserves QP feasibility under input constraints using a forward-simulated backup policy. Compared with a state-of-the-art 3DGS safety filter, our approach reduces trajectory jerk by 47% and runs 2.25 times faster. We validate the method in simulation and on hardware for real-time navigation in cluttered, perception-derived environments.
Problem

Research questions and friction points this paper is trying to address.

3D Gaussian Splatting
safety filter
quadrotor flight
actuator limits
realization gap
Innovation

Methods, ideas, or system contributions that make the work stand out.

3D Gaussian Splatting
Control Barrier Function
Quadrotor Dynamics
Safety Filter
Real-time Navigation