Research on Multi-agent Robot Behavior Learning Based on Fuzzy Neural Network

Human behavior or motion is diverse and complex. In order to design a robot with better sensitivity and better performance, it is necessary to set up multiple control nodes to facilitate the control robot to imitate the trajectory of human motion. However, due to the traditional robot behavior learning method, the control nodes are relatively single, and there is no clear behavior target node for the reference object, which leads to a large deviation in the robot behavior learning trajectory. Therefore, a fuzzy neural network-based Multi-agent robot behavior learning method. This method is based on the fuzzy neural network to control the behavior of robot. By optimizing the learning parameters of robot behavior, it can enhance the search program of behavior learning. According to the multi-level robot behavior learning model, it can identify the master-slave target of reference object and realize a more accurate multi-agent robot behavior learning method. The test results show that compared with the traditional method, the behavior trajectory of the proposed method is basically consistent with the behavior trajectory of the reference object. It can be seen that the method has better performance and meets the research requirements at the current stage.