Cooperative Formation And Hunting Of Multiple Security Robots In Cluttered Dynamic Environments

With the rise of the concept of "robot + security",security robots provide new technological approaches to intelligent security,which are of great significance in enhancing national public safety and counter-terrorism capabilities.Currently,there are widespread challenges regarding security robots operating in complex and dynamic environments,including insufficient multi-robot collaboration abilities.To address these issues,this paper focuses on the direction of formation control and path planning for security robots,specifically in the context of routine inspections.The aim is to enhance the inspection capabilities of multiple security robots in complex environments,addressing issues such as inadequate environmental adaptability and complex formation controllers.Additionally,an important research direction of this paper is improving the efficiency of apprehending suspicious targets when detected by security robots.The main work is as follows:In order to improve the flexibility and stability of multi-security robot systems in complex and dynamic environments,this paper proposes a hybrid path planning algorithm and formation control strategy under the leader-following formation structure.First,under the condition of a known environment map,an improved grey wolf algorithm based on whale hunting mechanism is designed to perform global path planning for the leader.When dynamic obstacles suddenly appear during the formation movement,a dynamic window approach is introduced to perform local path planning for the leader robot to avoid the dynamic obstacles.Based on this,a virtual leader robot position corresponding to the following robot is generated according to the leader robot’s position and formation parameters,and a trajectory tracking controller is designed to make the follower track the virtual leader robot’s motion to a predetermined position.The stability of the formation controller is proven using Lyapunov function.Finally,the simulation confirms the effectiveness and superiority of the proposed method.This paper proposes a multi-robot online cooperative capture strategy based on a buffer Voronoi diagram in a dynamic and complex environment,which is used when a suspicious target is detected during the security robot patrol process.Firstly,based on the buffer Voronoi diagram definition and the principle of support vector machines,the boundary weights between robots and robots,and robots and obstacles are dynamically updated to generate corresponding dynamic buffer Voronoi regions,ensuring that robots can move safely within this area without colliding with other robots and obstacles.Secondly,a capture encirclement is constructed for each escaped robot,and capture points are evenly distributed on the encirclement based on the number of capture robots,and a collaborative optimization capture allocation strategy is proposed for the capture robots and capture points based on the Hungarian algorithm to minimize the total tracking path of all capture robots.Finally,a capture controller based on collision risk is designed to improve the capture capability of robots and save capture time.Simulation results show the superiority of the proposed method in terms of capture time and travel distance.