SAFER-Splat: A Control Barrier Function for Safe Navigation with Online Gaussian Splatting Maps

2024-09-15

arXiv.org

AI Summary

To address safe navigation for robots operating in real-time, densely reconstructed Gaussian Splatting (GSplat) maps, this paper proposes the first scalable Control Barrier Function (CBF)-based filter tailored for massive Gaussian primitives. The method tightly integrates CBFs with online GSplat mapping, enabling low-overhead, minimally invasive real-time action safety filtering without disrupting GPU-accelerated rendering. A novel lightweight barrier constraint is designed to jointly ensure safety, computational efficiency, and appropriate conservatism. Simulation results demonstrate that our approach achieves 20–50× higher inference speed than NeRF-based methods, while providing stronger safety guarantees and more permissive control decisions. Real-world validation on an embedded UAV platform confirms simultaneous visual-inertial mapping and collision-free navigation using only onboard monocular vision—no external localization or pre-built maps required.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Natural Language Processing: Safety and Robustness
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Security and Privacy: Large-scale security measurements
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

SAFER-Splat (Simultaneous Action Filtering and Environment Reconstruction) is a real-time, scalable, and minimally invasive action filter, based on control barrier functions, for safe robotic navigation in a detailed map constructed at runtime using Gaussian Splatting (GSplat). We propose a novel Control Barrier Function (CBF) that not only induces safety with respect to all Gaussian primitives in the scene, but when synthesized into a controller, is capable of processing hundreds of thousands of Gaussians while maintaining a minimal memory footprint and operating at 15 Hz during online Splat training. Of the total compute time, a small fraction of it consumes GPU resources, enabling uninterrupted training. The safety layer is minimally invasive, correcting robot actions only when they are unsafe. To showcase the safety filter, we also introduce SplatBridge, an open-source software package built with ROS for real-time GSplat mapping for robots. We demonstrate the safety and robustness of our pipeline first in simulation, where our method is 20-50x faster, safer, and less conservative than competing methods based on neural radiance fields. Further, we demonstrate simultaneous GSplat mapping and safety filtering on a drone hardware platform using only on-board perception. We verify that under teleoperation a human pilot cannot invoke a collision. Our videos and codebase can be found at https://chengine.github.io/safer-splat.

Problem

• Safe robotic navigation using real-time Gaussian Splatting maps.
• Minimally invasive action filtering for collision-free robot movement.
• Efficient online mapping and safety control for autonomous drones.

Innovation

• Real-time safe navigation using Gaussian Splatting maps
• Minimally invasive action filter with control barrier functions
• Efficient GPU resource usage for online Splat training