一种基于嵌入式平台的机械臂体感控制系统

A mechanical arm motion sensing control system based on the embedded system

针对传统机械臂控制方式复杂且效率低的问题,研制了一套惯性式动作捕捉的体感控制系统.该系统在用户手臂绑定多个节点,每个节点由加速度传感器、陀螺仪和磁通传感器组成.节点采集手臂运动数据发送到嵌入式系统,经过融合计算得到手臂运动状态,并最终得出机械臂舵机转动角度.在保证机械臂控制精度的前提下,针对嵌入式系统硬件条件采用了降低系统延迟的分状态数据融合方案.本系统在嵌入式平台上实现,降低了动作捕捉成本,提高了可移植性.测试结果表明,使用惯性式动作捕捉的机械臂体感控制系统动作响应延迟时间短、机械臂动作较平稳.

The traditional mechanical arm control is complex and inefficiency,so a mechanical arm control system for human motion capture using inertial sensor is designed and implemented in this paper. The system uses sensor nodes,which contain accelerometer,gyroscope and magnetic sensor,to collect motion data of the arms and transmit to the embedded system. Through data aggregation,the quaternion which expresses the moving state of arm is calculated and finally the angle of mechanical arm＇s steering engines is obtained. Under the premise of ensuring the accuracy of mechanical arm,a data integration program of the different state is designed to reduce delay time for the embedded system,the system is implemented in an embedded system with low cost and portability. The experimental results show that the operating lag of system is short and the movement of mechanical arm is steady.