Multi-Agent Obstacle Avoidance using Velocity Obstacles and Control Barrier Functions (ICRA 2025)

Alejandro Sánchez Roncero, Rafael I. Cabral Muchacho, and Petter Ögren

---

This repository contains the MATLAB code used for the experiments in the paper "Multi-Agent Obstacle Avoidance using Velocity Obstacles and Control Barrier Functions."

Overview

This project provides a MATLAB implementation that:

• Uses Velocity Obstacles (VO) for guidance as a relaxed constraint in the optimization objective.
• Guarantees safety through a Control Barrier Function (CBF) constraint.
• Works with both second-order integrator dynamics and car-like dynamics.

There are two major folders—one for second-order integrator dynamics and another for car-like dynamics. Each folder has the same structure:

• h_cbf_system.m: Main file implementing the collision avoidance algorithm, including all methods and the optimization problem.
• run_scenario.m: Helper script to run simulations, with parameters specified in sim_config.m.
• scenarios/: Folder containing various scenarios. Each scenario is constructed as a Simulink file.

If you encounter any issues with the code or have suggestions for improvement, please create a pull request or email [email protected]. Note that the same method can be applied to more complex dynamical systems.

Requirements

• MATLAB
• Simulink and Simulink 3D Animation
• Symbolic Math Toolbox (used by generate_functions.m to create methods stored in generated_functions/. If you do not have this toolbox, you can implement those methods manually.)

How to Run the Code

1. Open MATLAB and set the current folder to the folder corresponding to the dynamics you want to simulate.
2. Configure sim_config.m with the desired settings for the experiments.
3. Run run_scenario.m to start the simulation.

Warnings

By default, all demo videos are recorded at a high frame rate, which can result in large file sizes.

Citation

If you use this code in your research, please cite our paper:

@misc{roncero2024multiagentobstacleavoidanceusing,
      title={Multi-Agent Obstacle Avoidance using Velocity Obstacles and Control Barrier Functions}, 
      author={Alejandro Sánchez Roncero and Rafael I. Cabral Muchacho and Petter Ögren},
      year={2024},
      eprint={2409.10117},
      archivePrefix={arXiv},
      primaryClass={cs.RO},
      url={https://arxiv.org/abs/2409.10117}, 
}