Distributed collaboration: Cognitive difference and collaborative decision for multi-robot radioactive source search

Accidental loss of radioactive sources can pose a significant threat to human health and the ecological environment. This paper proposes a distributed collaborative decision algorithm for multi-robot radioactive source search to solve this problem. The algorithm utilizes a Gaussian parameter fusion method based on cognitive consistency to estimate the posterior probability distribution of source parameters within a Bayesian framework. Also, the algorithm enhances the cognitive consistency of robots through the fusion of shared Gaussian distribution parameters and cognitive difference variables. Each robot calculates a reward function for its next action based on the deviation angle and distance. In addition, the proposed algorithm uses a distributed collaborative search strategy to guide the robot's search through the robot's behavioral decisions and behavioral decisions from neighboring robots. Experimental results show that the algorithm achieves a mean search time of only 1.74s and a 100% search success rate. Therefore, the algorithm not only guarantees accuracy but also shortens the search time and reduces the risk of losing radioactive sources.

Automated summary

This paper proposes a distributed collaborative decision algorithm for multi-robot radioactive source search, which estimates the posterior probability distribution of source parameters using cognitive consistency and Gaussian parameter fusion. Experimental results show that the algorithm achieves a high success rate and short search time, reducing the risk of losing radioactive sources.