A novel flocking search method of multiple agents

Most conventional cooperative search methods typically operate in an offline and centralized fashion, which inherently constrains their adaptability to dynamic environmental conditions. In response to this limitation, we propose a new flocking search method that leverages the novel local rules to coordinate and guide multiple intelligent agents to achieve a flocking state, efficiently cover the mission region and minimize environmental uncertainty. The local rules governing our flocking search methodology comprise both flocking rules and guiding rules. Flocking rules govern the interactions among agents are represented by a newly devised function that designed to maintain relatively stable distances between agents and facilitate smooth positional transitions at specified distances. Notably, this function accounts for the force reduction at both the closest and farthest distances between agents. Thereby, the local rules render the control law curve more practical and logical. Furthermore, we incorporate the information entropy of nearby grids surrounding each agent as the guiding rules.Finally, the simulations verify that proposed method can achieve the full coverage of the mission area efficiently and the searching efficiency of the proposed flocking search method outperform the traditional searching algorithms.