Remote Control And Implementation Of UAV Based On ROS And EEG

In recent years,the unmanned aerial vehicle(UAV)industry has developed rapidly and has shown broad application prospects in all walks of life due to its advantages of portability,flexibility,low cost and safety.At present,UAVs commonly used in the market are artificially remotely controlled,and people’s hands have not yet been completely liberated.Brain Computer Interface(BCI),as a new technology,can realize the interaction between brain and external devices.Stable State Visual Evoked Potential(SSVEP)is a kind of experimental paradigm of non-intrusive brain machine interface,which has been widely concerned and studied because of its high recognition accuracy and stable characteristic pattern.Combining the brain machine interface with the unmanned aerial vehicle to realize the UAV that only relies on the operator’s "independent consciousness",can accelerate the practical application of the brain control equipment,further expand the practical scope of the UAV,and provide greater convenience for the specific group and special task.In addition,in the current UAV development,most of the manufacturers use self-developed UAV platforms,which follow different standards and specifications.It increases the difficulty of transplantation,and causes developers to repeatedly develop the same functions,wasting a lot of time and effort.Robot Operating System(ROS)as a universal robot architecture,greatly improves the portability,code reuse and other issues.It supports multilingual programming.The operation and compilation is independent of the platform,and it is easy to use.Therefore,this paper combines BCI technology and ROS system to realize a modular,portable and highly integrated brain-controlled UAV system.This solution not only solves the problem of the expansion of the UAV function module and the system portability,but also realizes the real-time flight control of the human brain wave signal to the UAV.The main work of this paper is as follows:1.The existing problems in the development of traditional UAV are analyzed.The implementation scheme based on ROS system and EEG development is proposed,and the overall framework of the UAV system is given.The flight principle,dynamic model and ROS system framework of UAV are fully studied,which lays the foundation for the development of brain-controlled UAV.2.Combining the ROS system to complete the design of the system on the M100 machine,this paper develops the M100 UAV function module and designs the UAV control instruction and program,realizing the remote control of the UAV.The whole process of the system development is introduced in detail,the details of the design of the UAV instruction and the development of the control program are given.The C10 X module is used to realize the remote communication.The test results show that the wireless module can send the instructions and receive the feedback information correctly.The simulated flight test shows that UAV equipped with ROS system can send commands and respond instantly.3.This paper uses SSVEP-BCI to complete the construction of brain controlled UAV system.First,this paper designs the stimulus paradigm of SSVEP.Secondly,this paper studies the two algorithms of Canonical Correlation Analysis(CCA)and Filter Bank Canonical Correlation Analysis(FBCCA),and then makes a implementation.Then,this paper compares the offline experiments’ accuracy of CCA and FBCCA under different time windows.The experimental results show that with the increase of time windows,the accuracy of the two methods increases.Under the same time window,the accuracy of FBCCA recognition is higher than that of CCA recognition.Therefore,FBCCA is selected as the signal analysis algorithm for the brain-controlled UAV,and the time window of 1.5s is taken for online experiment.The results show that the average recognition accuracy of FBCCA can reach more than 90%,and the FBCCA algorithm can be fully competent for the control task of UAV in the accuracy rate.In this paper,the integration of UAV and EEG technology has been successfully applied to UAV control system,which has significant research and application significance in military field.