Motion Intent Recognition in Intelligent Lower Limb Prosthesis Using One-Dimensional Dual-Tree Complex Wavelet Transforms.

Min Sheng,Hui-Lin Chen,Ting-Ting Tong,Wan-Jun Wang,Yuan-Yuan Yang,Ben-Yue Su,Silvia Conforto

doi:10.1155/2021/5631730

Min Sheng, Hui-Lin Chen + Show 5 more

Open Access

https://doi.org/10.1155/2021/5631730

Copy DOI

Abstract

The motion intent recognition via lower limb prosthesis can be regarded as a kind of short-term action recognition, where the major issue is to explore the gait instantaneous conversion (known as transitional pattern) between each two adjacent different steady states of gait mode. Traditional intent recognition methods usually employ a set of statistical features to classify the transitional patterns. However, the statistical features of the short-term signals via the instantaneous conversion are empirically unstable, which may degrade the classification accuracy. Bearing this in mind, we introduce the one-dimensional dual-tree complex wavelet transform (1D-DTCWT) to address the motion intent recognition via lower limb prosthesis. On the one hand, the local analysis ability of the wavelet transform can amplify the instantaneous variation characteristics of gait information, making the extracted features of instantaneous pattern between two adjacent different steady states more stable. On the other hand, the translation invariance and direction selectivity of 1D-DTCWT can help to explore the continuous features of patterns, which better reflects the inherent continuity of human lower limb movements. In the experiments, we have recruited ten able-bodied subjects and one amputee subject and collected data by performing five steady states and eight transitional states. The experimental results show that the recognition accuracy of the able-bodied subjects has reached 98.91%, 98.92%, and 97.27% for the steady states, transitional states, and total motion states, respectively. Furthermore, the accuracy of the amputee has reached 100%, 91.16%, and 90.27% for the steady states, transitional states, and total motion states, respectively. The above evidence finally indicates that the proposed method can better explore the gait instantaneous conversion (better expressed as motion intent) between each two adjacent different steady states compared with the state-of-the-art.

Highlights

Introduction e2011 World Disability Report points out that there are at least 30 million amputees in developing countries [1,2,3,4]. e prosthesis can allow amputees to maintain the limb balance and compensate for the body appearance
We propose an improved motion intent recognition method based on the mechanical signals via intelligent lower limb prosthesis. e main contributions of this paper are listed as follows: (1) We introduce the one-dimensional dual-tree complex wavelet transform (1D-DTCWT) to study the transitional pattern between two adjacent different steady states so as to identify the motion intent of the lower limb amputees. e wavelet transform has the ability of the time-frequency local analysis, which can amplify the instantaneous variation characteristics of gait information, making the extracted features of instantaneous pattern more stable
The testing accuracy reached 100% ± 0.00%. e result may be related to the length of time the amputee subject wears the prosthesis. e steady states are such common movements in daily life, and it is very important for amputees to more accurately recognize them

Summary

Introduction

Introduction e2011 World Disability Report points out that there are at least 30 million amputees in developing countries [1,2,3,4]. e prosthesis can allow amputees to maintain the limb balance and compensate for the body appearance. E prosthesis can allow amputees to maintain the limb balance and compensate for the body appearance. It improves the amputees’ integration into society and restores their ability to works [5]. E motion intent recognition via the lower limb prosthesis requires identifying the gait instantaneous conversion (known as transitional pattern) between each two adjacent different steady states of gait mode. E sEMGs [8], which are collected by the biological electrode attached to the skin surface, can reflect the muscle contraction and relaxation. Huang et al [9] proposed a sEMGs-based human motion recognition method, which successfully recognized 7 daily motion states. The sEMGs can better reflect the strength of muscle contraction, they are affected by the nerve atrophy and electrode position [10]

Methods

Results

Conclusion