A Novel Distributed Hybrid Cognitive Strategy for Odor Source Location in Turbulent and Sparse Environment

Area of Science:

• Robotics
• Artificial Intelligence
• Environmental Science

Background:

• Accurate odor source localization is critical for robotic applications in hazardous environments.
• Turbulent and sparse conditions pose significant challenges to existing localization methods.

Purpose of the Study:

• To develop an advanced multi-agent framework for robust and efficient odor source localization.
• To enhance the performance of robotic systems in hazardous chemical monitoring and disaster response.

Main Methods:

• Cooperative Gravitational-Rényi Infotaxis (CGRInfotaxis): a distributed framework integrating multi-agent collaboration and hybrid cognitive strategy.
• Gravitational potential field for rapid source convergence and Rényi divergence for probabilistic exploration in sparse conditions.
• Particle filtering for posterior probability estimation and a distributed interactive-optimization mechanism for real-time cooperation.

Main Results:

• CGRInfotaxis achieved a near-100% success rate with high consistency across various scenarios (fixed/random sources, varying agents).
• The method demonstrated superior stability and adaptability compared to existing approaches.
• Increased agent numbers enhanced search efficiency without sacrificing reliability.

Conclusions:

• CGRInfotaxis significantly advances multi-agent odor source localization in turbulent, sparse environments.
• The framework offers practical utility for real-world applications in hazardous monitoring and disaster response.
• The cooperative, hybrid strategy provides a robust solution for complex environmental sensing tasks.