基于EEG能量解码的机械手控制方法研究

发布时间：2018-03-29 01:33

本文选题：数字舵机　切入点：插补算法　出处：《沈阳工业大学》2017年硕士论文

【摘要】：利用人类脑电波对机械手臂进行控制是机电工程和信息学领域所关注的重要问题。近年来,脑机接口技术取得重要进展,但较笨重的多导联脑电采集装置及相对复杂的控制方法限制了其在实际场景中的应用。为在一定程度上解决上述问题,本文探索了基于便携脑电传感器的EEG能量解码机械手控制方法,并在以下几方面取得进展。舵机的控制效果是机械手动作稳定输出的关键。为此,分析了数字伺服舵机模块的构成,工作原理和性能特点,构建了数字舵机的PROTEUS仿真驱动电路和实际硬件驱动电路,利用定时器技术设计了宽度可调的周期控制脉冲,研究了常用数字伺服舵机控制脉冲宽度和动作输出角度之间的线性匹配关系。为了克服常规的数字舵机控制机械手易出现动作僵硬,失调发热等问题,研究了基于定时插补的可调速舵机控制方法,设计了定时插补算法,编制了算法实现代码。把STC12C5A60S2设置成主控芯片,实现了对大扭矩数字舵机MG996R的调速控制。实际的机械手臂通常具有6个以上自由度,因此,对上述算法进行改进使其具有控制8路舵机同时工作的能力。常规的多导联脑电采集系统结构复杂,不便携带,和外控设备之间的有线连接限制了其实际应用。为克服上述问题,本研究以集成了TGAM模块的头带式传感器作为EEG采集装置,提取了额部EEG,以串行通信方式传入上位机,解析了脑电信号数据包所包含的专注度和放松度信号,并对专注度信号进行能量积分,作为机械手控制信号的输入信号源。同时,研究并实现了基于PS2通信协议的无线机械手控制方法,以克服有线连接在不便实际应用的问题。在上述工作的基础上,对机械手、机械臂、数字舵机、控制器和脑电信号采集系统进行了集成,尝试采用脑电信号进行处理并对机械手进行控制,探索了对本系统性能进一步改进方向和具体方法。
[Abstract]:The use of human brain waves to control robotic arms is an important concern in the fields of electromechanical engineering and informatics. In recent years, important advances have been made in brain-computer interface technology. But the relatively heavy multi-lead EEG acquisition device and the relatively complex control method limit its application in the actual scene. In order to solve the above problems to a certain extent, In this paper, the control method of EEG energy-decoded manipulator based on portable EEG sensor is explored, and progress is made in the following aspects. The control effect of the steering gear is the key to the stable output of the manipulator. The structure, working principle and performance characteristics of the digital servo steering gear module are analyzed. The PROTEUS simulation driving circuit and the actual hardware driving circuit of the digital servo steering gear are constructed, and the cycle control pulse with adjustable width is designed by using the timer technology. The linear matching relationship between the width of control pulse and the angle of motion output of digital servo steering gear is studied. The control method of adjustable speed steering gear based on timing interpolation is studied, the timing interpolation algorithm is designed, and the algorithm realization code is worked out. The STC12C5A60S2 is set as the main control chip. The speed control of large torque digital steering gear MG996R is realized. The actual mechanical arm usually has more than 6 degrees of freedom, so, The algorithm is improved so that it has the ability to control the 8-way steering gear working at the same time. The conventional multi-lead EEG acquisition system is complex in structure and inconvenient to carry. In order to overcome the above problems, the headband sensor integrated with TGAM module is used as the EEG acquisition device, the frontal part of EGG is extracted and transmitted to the upper computer by serial communication. The focus and relaxation signals contained in the EEG data packet are analyzed, and the energy integration of the focus signal is carried out as the input signal source of the manipulator control signal. The control method of wireless manipulator based on PS2 communication protocol is studied and realized in order to overcome the problem that wired connection is inconvenient for practical application. The controller is integrated with the EEG acquisition system, and the electroencephalogram (EEG) signal is used to process and control the manipulator. The further improvement direction and concrete method of the system performance are explored.
【学位授予单位】：沈阳工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP241

【参考文献】