Portable Hybrid Human-machine Interface System And Its Application In Intelligent Driving

In recent years,with the development of human-machine interface control technology based on bioelectrical signals in the fields of rehabilitation equipment and therapeutic equipment,it has become a new research hotspot in the field of human-computer interaction such as voice control and visual perception.In order to help people with movement disorders improve their autonomy in life and reduce their dependence on medical staff,this thesis proposes a portable multi-modal human-machine interface that can help people with movement disorders control the driving process of vehicles and pass in barrier-free areas.The control of the intelligent driving system is realized by contraction of the left and right masticatory muscles of the facial muscles,and the temporal muscle EMG signals collected in the scalp of the temporal lobe of the head,which are processed by the system to control the lateral(left and right)movement of the smart car,And use the EOG signal(EOG)generated by eye movements to control the start-stop movement of the vehicle.The advantages of this system are that,first,the EEG signal acquisition channel is used to collect EMG signals,which can concentrate all the electrodes on the human head,which is more convenient than collecting EMG signals directly.The second is that the EMG has a wide bandwidth,strong anti-environmental interference ability,and simpler processing than EEG signals;at the same time,collecting blink signals on the forehead has the characteristics of high signal-to-noise ratio and stable waveform.The third is the continuous control output of the EMG signal generated by the tooth occlusion in the temporal lobe area.Combining existing related technologies,in order to realize the system design proposed in this thesis,the main work of this thesis is as follows:First,the design and implementation of the human-machine cooperative driving system,including the design of the collection device of myoelectricity and oculogram,the design of signal processing modules,the modeling and analysis of the human-machine cooperative vehicle steering system,the design of assisted driving system,and the implementation of each module System debugging and programming,etc.Secondly,verify the system’s offline classification model and analyze the stability and validity of online experimental results.Offline experiments mainly carry out the verification of the algorithm model and the acquisition of EMG and EOG signal parameters,and select the algorithm model with better effect and easy to implement for the online experiment of the system;use the online experiment to check the stability and effectiveness of the system Verification and statistics of relevant performance parameter information and comparison with existing system performance parameters.The system in this thesis has an average classification accuracy rate of 88.32% for EOG and EMG signals,and an average error operation rate of 0.25 times/min.It can only initially have the function of human-machine cooperative driving.In future work,in order to adapt to more users,especially patients with movement disorders,more data of the subjects and further optimization of the system are needed.