# Safe Quadrotor Navigation using Composite Control Barrier Functions

**Authors:** Marvin Harms, Martin Jacquet, Kostas Alexis  
**arXiv:** [2502.04101](https://arxiv.org/abs/2502.04101)  
**Date:** 2025-02-06  
**Summary page:** https://www.summarizepaper.com/en/arxiv-id/2502.04101v1/

## Key points

- Authors: Marvin Harms, Martin Jacquet, and Kostas Alexis
- Introduction of a safety filter for collision avoidance in multirotor aerial robots
- Utilization of a single Composite Control Barrier Function considering all position constraints in third-order nonlinear representation
- Analysis of recursive feasibility demonstrating negligible infeasible set under composite constraint
- Computational scalability against thousands of constraints for navigation in complex environments with numerous obstacles
- Experimental validation showcasing effectiveness in guaranteeing quadrotor safety equipped with LiDAR system indoors and outdoors
- Testing against naive and adversarial nominal policies highlighting robustness and reliability
- Potential for enhancing collision avoidance capabilities in autonomous aerial systems
- Accepted for presentation at ICRA 2025, contributing valuable insights into improving quadrotor navigation safety and efficiency through advanced control barrier functions

## Summary

In their paper titled "Safe Quadrotor Navigation using Composite Control Barrier Functions," authors Marvin Harms, Martin Jacquet, and Kostas Alexis introduce a safety filter designed to ensure collision avoidance for multirotor aerial robots. The proposed formalism utilizes a single Composite Control Barrier Function that considers all position constraints within a third-order nonlinear representation of the robot's dynamics. Through an analysis of the recursive feasibility of the safety filter under the composite constraint, the authors demonstrate that the infeasible set is negligible. This innovative method enables computational scalability against thousands of constraints, allowing for navigation in complex environments with numerous obstacles. The experimental validation conducted by the authors showcases the effectiveness of their approach in guaranteeing the safety of a quadrotor equipped with an onboard LiDAR system. The experiments were carried out in both indoor and outdoor cluttered environments, testing the quadrotor's ability to navigate safely against both naive and adversarial nominal policies. The results highlight the robustness and reliability of the proposed safety filter in real-world scenarios, emphasizing its potential for enhancing collision avoidance capabilities in autonomous aerial systems. Accepted for presentation at the International Conference on Robotics and Automation (ICRA) 2025, this research contributes valuable insights into improving the safety and efficiency of quadrotor navigation through advanced control barrier functions. By addressing key challenges related to collision avoidance in dynamic environments, this work paves the way for enhanced autonomy and reliability in aerial robotics applications.