CAJ | 학술논문

在假肢运动优化控制的研究中，针对动力型假肢控制方面存在的运动模式识别准确性差的问题，搭建人体下肢运动信息系统获取下肢髋关节角速度信号和加速度信号。建立基于BP神经网络的大腿截肢者运动识别模型。研究了建模过程中输入输出数据预处理、网络结构设计、训练模式选择等问题。改进模型能有效识别平地行走、上楼、下楼、上坡和下坡5种运动模式，正确识别率达到了90.4%，已具备一定的实用性。
재가지운동우화공제적연구중，침대동력형가지공제방면존재적운동모식식별준학성차적문제，탑건인체하지운동신식계통획취하지관관절각속도신호화가속도신호。건립기우BP신경망락적대퇴절지자운동식별모형。연구료건모과정중수입수출수거예처리、망락결구설계、훈련모식선택등문제。개진모형능유효식별평지행주、상루、하루、상파화하파5충운동모식，정학식별솔체도료90.4%，이구비일정적실용성。
Lower limb amputation significantly affects the quality of the leg amputee's daily life. Recent advance-ments in electromechanical actuators have propelled the recent development of powered artificial legs. Accurately rec-ognizing the leg amputee's locomotion intent is required in order to realize the smooth and seamless control of prosthet-ic legs. The approach infers amputee's intents of upslope, downgrade, stairs ascent, stairs descent or level-ground walking without the need for instrumentation of the sound-side leg. Specifically, the intent recognizer utilizes the fea-tures extracted from accelerometer and gyroscope. The preprocessing of input and output data, design of network structure, training mode selection and other aspects were analyzed. This paper demonstrates via experiments the ef-fectiveness of the approach.