Sailing Through Point Clouds: Safe Navigation Using Point Cloud Based Control Barrier Functions

The capability to navigate safely in an unstructured environment is crucial when deploying robotic systems in real-world scenarios. Recently, control barrier function (CBF) based approaches have been highly effective in synthesizing safety-critical controllers. In this letter, we propose a novel CBF-based local planner comprised of two components: Vessel and Mariner. The Vessel is a novel scaling factor based CBF formulation that synthesizes CBFs using only point cloud data. The Mariner is a CBF-based preview control framework that is used to mitigate getting stuck in spurious equilibria during navigation. To demonstrate the efficacy of our proposed approach, we first compare the proposed point cloud based CBF formulation with other point cloud based CBF formulations. Then, we demonstrate the performance of our proposed approach and its integration with global planners using experimental studies on the Unitree B1 and Unitree Go2 quadruped robots in various environments.

I. Introduction

Mobile robots have been deployed to navigate unstructured environments autonomously for various operations [1], [2], [3], [4], [5], [6], [7], [8]. Traditionally, work has been done in generating motion plans for navigation using pre-built maps with global motion planners like rapidly exploring random trees (RRTs) [9]. In [10], a safe corridor based approach is used to reduce the safe set to polytopes along a pre-planned path, which enables online replanning. However, when the map has high granularity as well as when the map is constructed/updated in real-time, it would be costly to frequently replan using these global planners, creating the need for a local planner that directly operates on sensor measurements while ensuring safety. Recently, control barrier functions (CBFs) [5] have been gaining popularity in synthesizing safe control actions. One of the main benefits of CBF-based approaches is that they can transform nonlinear and nonconvex constraints into linear ones, greatly increasing the computation speed.