The Research About The Decision System Of Multi-robots Navigation Behavior Based On The Emotion Apperceive

As an artificial system, Multi-robots system is actually a simulation of the group system of nature and human society. The basic idea of Multi-robots coordination and control is to consider multi-robots system as a group or a society and to study the coordination mechanism between multiple robots from the perspective of organizations and systems, so as to give full play to various inherent advantages of the multiple robot system.Navigation is one of important research contents of mobile robots. The navigation of multiple mobile robots does not only include robot working environment modeling, path planning and robot motion control, but also include the research of communication and information between multi-robots. Currently, the single robot navigation technology under the global known environment has tended to be mature. However, taking effective implementation of multi-robot navigation under the complex unknown environment is still a pressing problem. Aiming at this kind of problem, this paper solves for the robot work environment modeling, path planning and information exchange of the three key issues, so it has great theoretical significance and application value. The major work carried out as follows:1) In the system, multi-robots system use lead-follow multi-robots cooperative navigation structure based on priority. That is, every robot makes behavior decision and judges the environment around themselves through wireless communication. Then, the robots can acquire own priority according to the environment conditions. The robot which has the highest priority is the temporary lead-robot which has priorities for action and commanding other follow-robots. In this way, robots don't only have the ability of making decision independently but also make the groups not too dispersed, which ensures that the robot can complete the task efficiently and accurately.2) This system introduced the concept of emotion apperceive to describe the influence of obstacles in the environment on the robots. The learning and decision model based on emotion apperceive is a new reinforcement learning algorithm. The model make inner incentives from emotion and external incentives from apperceive as a motivation of learning and decision. The model is introduced to the multi-robots navigation decision system and constructs a new mobile robot navigation control system. With emotion apperceive learning based on online, the formation of multi-robots cooperation mechanisms can enhance the adaptability of mobile robot navigation in unknown environment, so that, the robot can complete autonomous navigation tasks fleetly and efficiently.3) This system introduced the concept of will strength to describe the influence of target point on the robots. If the robot is moving toward the target, the will strength of the robot on its moving direction is greater. If not, is smaller. The behavior of the robot toward the goal becomes clear and direct.The simulation results show that the algorithm researched by the subject realizes multi-robots autonomous navigation and increases the concept of emotion apperceive and will strength, which makes the system efficient, accurate and humanization.In the final part, we present discussion on the potential applications and the comparison between different navigation schemes. Conclusions and some possible improvements of the proposed navigation schemes are also presented.