Development of lower limb motion detection based on LPMS
Tongyang Sun,Chunbao Wang,Quanquan Liu,Zhijiang Lu,Lihong Duan,Pengfang Chen,Yajing Shen,Meng Li,Weiguang Li,Qihong Liu,Qing Shi,Yulong Wang,Jian Qin,Jianjun Wei,Zhengzhi Wu
DOI: https://doi.org/10.1109/RCAR.2016.7784033
2016-01-01
Abstract:Up to now, with the increasing of the elderly population, more and more patients are suffering from hemiplegia. It leads to a great need for hemiplegic rehabilitation. In traditional rehabilitation, each patient must be treated by therapist, one by one. However, since the individual differences of therapists, no effectiveness rehabilitation is guaranteed. And the rehabilitation status of patient is still diagnosed by therapists with their subjective experience. This would cause the inhomogeneity on rehabilitation evaluation and sometimes negative influence on the rehabilitation effect. To solve these problems, many research groups proposed rehabilitation evaluation systems to assess the status of the hemiplegic patients quantitatively. Rehabilitation motion detection is the basis of the evaluation system, and it requires the participation of therapist. However, many motion detection methods do not meet the detection requirements, such as mechanical tracking and optical sensor, etc. In this article we present a method to detect lower limb motion of hemiplegic patients based on inertial sensor technology. LPMS, a high performance, easy wearable, portable and large measurement range sensor, is selected as the motion sensor. We obtain the gesture quaternion of lower limb through LPMS, and then use the algorithm to convert quaternion to matrix and Euler angle. Combining with the simplified lower limb motion model, we compute the rotation angle of joint by processing the rotation quaternion in Matlab. Finally, the curve of rotation angle of knee is established. The method detecting the motion of lower limb can be integrated into the rehabilitation robot control system, realizing intelligent detection and evaluation. Thus, the rehabilitation robots could be expected adjusting training parameters based on patient status automatically, expected to have significant impacts in medical rehabilitation robot field.