Research On Brain-computer Cooperative Control Of Mobile Robot Based On Petri Nets

Brain-Computer Interface(BCI)is a new way of human-computer interaction,which can control external devices directly through the brain without using language and physical movements.Brain-controlled robot is an important research direction of BCI.The brain-controlled robot system combines the BCI system and the robot system,and collects the user's EEG signals to control the robot in real time,so that it can safely and efficiently perform the tasks expected by the user.At present,brain-controlled robots mainly have two control methods: direct control and shared control.Direct brain control has the disadvantages of low control efficiency and easy user fatigue.The shared control method can effectively improve the control efficiency of the brain-controlled robot and reduce the user's thinking burden.Therefore,the shared control method is becoming the main control method of the brain-controlled robot.The brain-computer cooperative control method based on Petri net proposed in this paper is a man-machine shared control method.The research focus of the paper will be developed in the following way:1.A brain-controlled robot system was established.The system consists of a BCI system and a robot system.The BCI system collects the 19-lead EEG signal of the user through the EEG amplifier,and then performs preprocessing,feature extraction,and classification operations on the collected EEG signal.According to the principle of two-level classification of the EEG signal of motor imaging,you can get straight,left,and right The EEG control commands are then designed for the AmigoBot mobile robot system to execute the above control commands.The brain-controlled robot system can realize direct brain control of the robot,and provides an experimental platform for the research of brain-computer cooperative control methods.2.Aiming at the shared control problem in BCI,the Petri net method is used for modeling.Combining the shared control strategy with the internal and external information of the BCI system,constructing a Petri net model of the BCI system for motion imaging,describing the operating status of the system,and facilitating the analysis and improvement of the shared control strategy In the Petri net modeling,a control place is added to describe the external input of the system,and the Petri net modeling method is simplified by synthesizing the Petri net method.In completing the Petri net modeling of the brain-controlled robot system,a Petri net model conversion algorithm was proposed,which could easily and safely implement the mutual conversion between the Petri net graphical model and the association matrix.Finally,it is verified by reachability graph that the model can correctly evolve the operation process of BCI system.3.Combine BCI system,Petri net system and AmigoBot robot system into brain-computer cooperative control system,and offline and online experiments are designed based on this system.The experimental results verify the proposed Petri net-based mobile the effectiveness of robot brain-computer cooperative control method.Compared with other shared control methods,the brain-computer cooperative control method of mobile robots based on Petri nets proposed in this paper can not only realize shared control between humans and machines,but also based on the dynamic evolution of Petri net models and reachability graph analysis,can be used to design and verify shared control strategies and help improve the applicability of BCI systems.