Research On Multi-robot Cooperation Pursuing Based On Local Control Strategy

To some extent, multi-robot systems represent the forefront of high-tech developmentwhich combines development results of multidisciplinary. A variety of intelligent robotsachieve self-moving oriented target and complete a certain task functions by which will beable to perceive the environment and their own state through the sensor, in order to worktogether thus by mutual cooperation and mutual coordination. In this paper, in the case of“node failure” led to hunting failure for pursuit-escape issues of multi-robot systems under thelocal control policy of distributed network, and the preliminary study is shown:First, this paper introduces the mutual cooperation process for pursuit-escape issues ofmulti-robot systems, distributed hunting is based on the position information between thehunting robots and target, because of environmental factors, such as obstacles, limitations ofperceived capacity and so on, and then consider if the target robot take certain escaping strategy,the target position information is lost during the hunting robot process and eventually leading tothe occurrence of the hunting failure situation.Secondly, the loss of the objective position information leads to hunting failure forcooperative hunting by swarm robotic systems, and the objective escapes on its own initiative,distributed hunting control (DHC) algorithm of combining cyclic pursuit strategy is proposed.The algorithm establishes the relationship among hunting robots and avoids the influencecaused by the loss of the objective position, the objective is captured successfully, then, cyclicpursuit strategy is improved, dynamic circle hunting formation is quickly formed based ontaking full advantage of the relative position relationship between the hunting robots andadjacent robots. Experimental results show that this algorithm not only can achieve thedesired hunting formation, but also the hunting consuming time is reduced greatly.Finally, the loss of the objective position information led to low efficiency forcooperative hunting by swarm robotic systems which combined cyclic pursuit strategy,distributed hunting control (DHC) algorithm based on topology was proposed. The algorithmdetermined the location of the virtual collaborative hunting robot according to the topology ofthe hunting system, and took full advantage of the position relationship of the hunting robots,then, the trajectory deviation between the invalid hunting robots and adjacent robots wasdecreased, dynamic circle hunting formation was formed quickly. Simulation resultsdemonstrate the correctness this proposed methods.