Discussion: Enhancing Cooperative Multi-Agent Mapping with e-Puck Robots and YOLOv8

[Yasouimo]

Hello everyone 👋,

We’re excited to share our project: “Cooperative Multi-Agent Mapping and Surveillance using e-Puck Webots Robots” – a collaborative initiative between Bellmir Yahya and Chegdati Chouaib, supervised by Pr. Hajji Tarik at ENSAM Meknès.

🔍 Project Overview:
Our robots operate within the Webots simulation environment using a 4-state navigation system, synchronized file-based communication, and YOLOv8 for real-time object detection. The system is capable of:

Distributed mapping of maze environments

Inter-robot communication and data sharing

Intelligent detection & alarm system (e.g., stray cats trigger alerts)

Real-time image processing with high-performance object detection

📊 YOLOv8 was chosen after benchmarking against other models due to its balance of speed and accuracy, essential for real-time robotics.

🔗 GitHub Repository: https://github.com/Yasouimo/Multi-agent-Mapping-and-Surveillance-Using-Webots-Bellmir-Chegdati

We’re opening this discussion to:

Gather feedback and suggestions on optimization (e.g., communication efficiency, mapping accuracy)

Exchange ideas around decentralized coordination in multi-agent systems

Hear from others working with Webots, YOLO, or multi-agent AI

💡Have questions, suggestions, or similar projects? Let’s connect and collaborate!

Looking forward to your insights!

— Bellmir & Chouaib