Research On Multiple Robots Coordination Exploration Algorithm In Indoor Environment

Multi-robot cooperation is a popular topic in robotics.A team of robots,though may be not powerful individually,can effectively compensate and compromise their limitations by cooperation.Multi-robot systems can significantly improve the efficiency or complete the tasks which the single robot can not complete.Mobile robots generally require a map to operate in a particular environment.Such operations include foraging,search and rescue,supervision and monitoring,surveillance,and navigation.The exploration mission of the unknown environment has a decisive effect on the completion of the task.The studies include the global map merging,the exploration tasks allocation,the avoidance of collisions among robots,and ability to detect obstacles in this paper.The specific work is as follows:Firstly,the global map merging algorithm is studied.It is divided into two parts according to the global map before and after merging.Before the map merging,each robot performs the Simultaneous Localization and Mapping,SLAM)task.Because of the computational complexity in the existing matching algorithm for hierarchical simultaneous localization and mapping,a signature vector based matching approach was proposed in this paper.The approach is improved by combining a landmark localization technique to realize more reliable loop closing calculation.In addition,the constraints equation was discussed for the case that two or more closed loops sharing a common sub-path for obtaining a global optimized map of the environment.In order to solve the repeat exploration problem,we proposed the atlas based mapping algorithm,which including the map division,the establishment of the topological node coordinate system,based on signature vector repeated exploration to avoid and so on.Simulation expeiments show that the map merging algorithm based on atlas has the advantages of short exploration time and low exploration repetition rate compared with the map merging based on encounter and the grid map method.Secondly,the emotion and CFC(clustering frontier cell)based bidding exploration coordination algorithm was proposed in this paper.In order to solve the problems existing in bidding-based coordination algorithms,the solutions are proposed from several aspects including exploration efficiency,optimal target selection,and exploration of isolated islands.The coordination algorithms based on bidding are most widely used in exploration taskcoordination,which the robot wins the bid with the best bid to explore the task.This kind of coordination algorithm produce much repeat path,so the explore efficiency is reduced.By introducing the emotional state of the robot in the bid based coordination algorithm,the repetition of exploring path can be effectively reduced.In order to overcome the problem of a non-optimal target being selected in the exploration process,a revised single linkage clustering frontier cell algorithm.Moreover,there exist unexplored islands for most of the bidding-based multi-robot coordination algorithms in the exploration of unknown environments.To deal with this problem,some rules switched by emotion states are proposed.We tested and evaluated our approach by means of experiments in a simulated environment and in real environment.The experiments results demonstrate that the exploration efficiency is improved,and our algorithm has a greater coverage rate than classic bidding-based coordination algorithms.Thirdly,the emotion and walkding rules based bidding exploration coordination algorithm is proposed in this paper.The algorithm has certain ability to recognize obstacles in the process of exploration.The emotional state generator based on Markov emotion model was put forward by studing the emotion models.The walking rules were given for avoiding the statics or dynamic obstacles,and the convex or non-convex obstacles in details.So,we propose the simulation results demonstrate that the emotion and walking rules based bidding exploration algorithm has a lower overlap rate than the bidding-based coordination algorithm.Thus,the exploration time is efficiently reduced.Moreover,the algorithm has a good ability of detecting static and dynamic obstacles.In the end,a strategy for collision avoidance for multi-robot system based on benefit is proposed in order to solve the collision problem between robots in performing the exploration task in this paper.The collisions among robots often occur when a multi-robot system performing exploration tasks.However,the collisions avoidance is different from the general obstacle avoidance.The obstacles generally are not moving in obstacle avoidance problem.The strategy of collision avoidance we propose determining the order of passing the intersection bases on the benefit of the robots,mainly focuses on improving the efficiency of multi-robot system.The situation of dynamic collision avoidance coordination is considered and the algorithm of determining the order of passing the intersection is given.Then,a simulation example of robots avoiding collision is given for illustrating the process of thestrategy.The results of the simulation are analyzed,and the reasonable assumption is given for the spatial relationship between the robots and targets.The simulation example verifies the feasibility of this method,and compared to the collision avoidance coordination algorithm based on the intersection of no traffic lights model,our strategy better solve the conflict problem of multiple robot through the intersection.