Research On Indoor Multi-robot Collaborative Search Metho

In recent years,the attention of researchers have been attracted to multi-robot system which is a part of the field of robotics research and has been widely used in military reconnaissance,resource detection,and hazardous environment inspections and rescues.Multi-robot collaborative target search is a hot topic in multi-robot research.Aiming at the problem of the efficiency and time of multi-robot target search in an unknown indoor environment,swarm intelligence algorithms is used to solve the problem of multi-robot collaborative search.The main tasks completed are as follows:First,the research background and significance of the multi-robot system are explained,and the domestic and foreign research status of the multi-robot collaborative search related technology is analyzed,as well as the existing problems and challenges.The mathematical model of the multi-robot target search problem is established,the basic idea of the particle swarm algorithm and the related concepts of the multi-robot cooperative search are analyzed,and the rationality of applying the particle swarm algorithm to the multi-robot cooperative search is analyzed.Then,aiming at the static single-object search problem in an unknown indoor environment,a multi-group fusion algorithm based on particle swarm optimization and longhorn beetle optimization algorithm is proposed.The algorithm draws on the idea of multi-group fusion,expands the search range of the multi-robot system,and improves the success rate of multi-robot collaborative search.It is verified by simulation that the algorithm is superior to traditional target search algorithms in terms of target search success rate and search efficiency.Then,in view of the influence of environment on target search success rate,a multi-robot cooperative search method based on target probability map is proposed.The algorithm uses the target signal model in the environment and the target signal information obtained by the robot to establish a target probability map.Through the designed multi-robot collaborative search strategy,the target point to move at the next moment is selected,and at the same time in the case of a large range of indoor obstacles(Such as walls,etc.),use the D*Lite algorithm to reach the target point selected by the search strategy,which reduces the influence of the environment on the success rate of target search.Finally,an indoor multi-robot collaborative search experiment platform is built.The platform is based on the Ardent framework of ROS2,which improves the reliability of multi-robot collaboration.With the help of the existing robot navigation and positioning function package,the multi-robot collaborative search algorithm proposed in this paper is tested and verified.The experimental results show that using the multi-robot cooperative search algorithm based on the target probability map,multi-robots can quickly and accurately search for the target.