Yuepeng Zhang

Dr. Yuepeng Zhang

orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 November 2023 in IEEE Sensors Journal
Verified
Journal Article
Closed access
LLLing-Long Li
Ling-Long Li
Guang-Zhong CaoGuang-Zhong Cao
Guang-Zhong Cao
HLHong-Jie Liang
Hong-Jie Liang
YZYuepeng Zhang
Yuepeng Zhang
FCFang Cui
Fang Cui
Human motion intention (HMI) has increasingly gained concerns in lower limb exoskeletons (LLEs). Human motion intention recognition (HMIR) is the precondition of realizing active compliance control in LLEs. Accurate and efficient recognition of the HMI will benefit the LLEs achieving natural and effective human-robot interaction (HRI) and improving the wearing comfort level. A systematic review on HMIR is of great significance in developing LLEs. However, there is no literature comprehensively describing the development roadmap of the human lower limb motion intention recognition (HLLMIR) in the LLEs so far. In order to have a comprehensive understanding of the HLLMIR and explore the current research status and development trend of LLEs, this paper provides a systematic review of the HLLMIR research for LLEs. Firstly, the HMI mechanism and understanding are fully illustrated, and the HMIR tasks pertaining to lower limb motions (LLMs) are elaborated on. Next, the intention-related sensing signals with different sources are dissected in detail, including bioelectric signals of electroencephalography (EEG) and electromyogram (EMG), biomechanical signals, and multisource signals fusion. The HMIR methods for the LLEs are thoroughly addressed and analyzed, the methods are categorized as model-based like musculoskeletal model, and model-free method involving Heuristic rule-based, conventional machine learning (ML)-based, and deep learning (DL)-based. Finally, an overall discussion on the recognition tasks, sensing signals, recognition methods, and performance assessments is given, and thus the research challenges of the HLLMIR are summarized and prospected.

17 January 2024 in IEEE Sensors Journal
Verified
Journal Article
Closed access
YZYuepeng Zhang
Yuepeng Zhang
Guang-Zhong CaoGuang-Zhong Cao
Guang-Zhong Cao
LLLing-Long Li
Ling-Long Li
DDDong-Feng Diao
Dong-Feng Diao
The interactive control of lower-limb exoskeleton robots (LLERs) is important to achieve compliance and safety. Significant challenges in the interactive control of LLERs include how to accurately recognize human motion intention, how to accurately recognize the motion environment, and how to achieve natural, stable, and accurate interactive control in accordance with the human motion intention and the environment. This paper presents a detailed classification of motion intention recognition, interactive control based on motion intention and interactive control based on the LLER’s motion environment and summarizes the methodologies in each category. In addition, the advantages and disadvantages of motion intention recognition, motion environment recognition and interactive control based on the motion intention and motion environment are analyzed from a macroscopic perspective, and current trends are discussed. Finally, this paper explores which requirements should be met in future LLER applications to improve the naturalness, stability, and accuracy of interactive control.

10 March 2025 in Sensors
Verified
Journal Article
Open access
PDF file
YZYuepeng Zhang
Yuepeng Zhang
JWJun Wu
Jun Wu
BGBo Gao
Bo Gao
LXLinzhong Xia
Linzhong Xia
忱卢忱 卢
忱 卢
Guang-Zhong CaoGuang-Zhong Cao
Guang-Zhong Cao
Manipulator robots hold significant importance for the development of intelligent manufacturing and industrial transformation. Manufacturers and users are increasingly focusing on fault diagnosis for manipulator robots. The voltage, current, speed, torque, and vibration signals of manipulator robots are often used to explore the fault characteristics from a frequency perspective, and temperature and sound are also used to represent the fault information of manipulator robots from different perspectives. Technically, manipulator robot fault diagnosis involving human intervention is gradually being replaced by new technologies, such as expert experience, artificial intelligence, and digital twin methods. Previous reviews have tended to focus on a single type of fault, such as analysis of reducers or joint bearings, which has led to a lack of comprehensive summary of various methods for manipulator robot fault diagnosis. Considering the needs of future research, a review of different fault types and diagnostic methods of manipulator robots provides readers with a clearer reading experience and reveals potential challenges and opportunities. Such a review helps new researchers entering the field avoid duplicating past work and provides a comprehensive overview, guiding and encouraging readers to commit to enhancing the effectiveness and practicality of manipulator robot fault diagnosis technologies.

3 November 2021 in Applied Sciences
Verified
Journal Article
Open access
PDF file
YZYuepeng Zhang
Yuepeng Zhang
Guang-Zhong CaoGuang-Zhong Cao
Guang-Zhong Cao
WLWenzhou Li
Wenzhou Li
JCJiangcheng Chen
Jiangcheng Chen
LLLing-Long Li
Ling-Long Li
DDDongfeng Diao
Dongfeng Diao
Lower limb rehabilitation exoskeleton robots have the characteristics of nonlinearity and strong coupling, and they are easily disturbed during operation by environmental factors. Thus, an accurate dynamic model of the robot is difficult to obtain, and achieving trajectory tracking control of the robot is also difficult. In this article, a self-adaptive-coefficient double-power sliding mode control method is proposed to overcome the difficulty of tracking the robot trajectory. The method combines an estimated dynamic model with sliding mode control. A nonlinear control law was designed based on the robot dynamics model and computational torque method, and a compensation term of control law based on double-power reaching law was introduced to reduce the disturbance from model error and environmental factors. The self-adaptive coefficient of the compensation term of the control law was designed to adaptively adjust the compensation term to improve the anti-interference ability of the robot. The simulation and experiment results show that the proposed method effectively improves the trajectory tracking accuracy and anti-interference ability of the robot. Compared with the traditional computed torque method, the proposed method decreases the tracking error by more than 71.77%. The maximum absolute error of the hip joint and knee joint remained below 0.55° and 1.65°, respectively, in the wearable experiment of the robot.

15 December 2022 in IEEE Sensors Journal
Verified
Journal Article
Closed access
ZLZi-Qin Ling
Zi-Qin Ling
JCJiang-Cheng Chen
Jiang-Cheng Chen
Guang-Zhong CaoGuang-Zhong Cao
Guang-Zhong Cao
YZYuepeng Zhang
Yuepeng Zhang
LLLing-Long Li
Ling-Long Li
SCSheng-Bin Cao
Sheng-Bin Cao
Gait phase recognition based on surface electromyography (sEMG) signals provides a human-robot interface for rehabilitation robots. However, a high-performance deep learning model requires to be calibrated on a large amount of data and frequently recalibrated for new gait speeds, which raises a burden on users. In this paper, a domain adaptive convolutional neural network (DACNN) model for sEMG-based gait phase recognition against speed changes is proposed, which only needs to be pre-trained on a comfortable gait speed and quick recalibrated on new gait speeds. Specifically, a CNN-based backbone model (BM) is first constructed and pre-trained using comfortable gait speed data. Then, a Domain Adaptation (DA) algorithm is applied for recalibration, which allows the BM to learn speed-invariant representation with a small amount of new speed data. Four subjects participated in the experiment and were required to walk at four gait speeds, three BMs based on AlexNet, LeNet and FDCNN were set as architectural options. Compared with the fine-tuning (FT) and non-recalibration (NoRC) strategies, the results show that the proposed method significantly outperforms both baseline methods. By recalibrating on 6s data, the proposed method can achieve an accuracy improvement of 1.5% to 6.1% compared to FT, and 58% to 81% compared to NoRC. The average gait phase recognition accuracies are 58.13% for AlexNet, 81.56% for LeNet, and 81.53% for FDCNN, respectively. The average mean absolute errors for gait event identification are 48ms, 85ms, and 66ms, respectively. These results indicate that DACNN with low recalibration burden can improve the usability of sEMG-based pattern recognition systems.

21 September 2022 in IEEE Sensors Journal
Verified
Journal Article
Closed access
ZLZi-Qin Ling
Zi-Qin Ling
YZYuepeng Zhang
Yuepeng Zhang
Guang-Zhong CaoGuang-Zhong Cao
Guang-Zhong Cao
JCJiangcheng Chen
Jiangcheng Chen
LLLing-Long Li
Ling-Long Li
DTDong-Po Tan
Dong-Po Tan
Step length estimation is essential for people to overcome different types of environmental obstacles in daily life. Aiming at the problem of low estimation accuracy based solely on surface electromyography (sEMG), in this paper, a multi-source (MS) data fusion method based on auto-encoder-convolutional neural network (AE-CNN) is proposed to estimate the motion parameters of human lower limbs, which improve the accuracy of step length estimation. Specifically, the time-domain feature data of sEMG and insole pressure data from the lower limbs are first collected simultaneously to form the MS. Then, by applying AE as the fusion units, fused data are obtained and dimensionality reduction is realized. Finally, by integrating AE and CNN, the step length can be well estimated by utilizing the MS. For experimental validation, the performances of different data fusion units and regression units in step length estimation are compared. The results show that the proposed method can achieve good estimation. The normalized root mean square error (NRMSE) and Pearson correlation coefficient (PCC) of proposed method reach 0.0479±0.0263 and 0.8273±0.1082, respectively. The NRMSE of the best trial is 0.0217, the PCC is 0.9774. Further, experiments at different walking speeds are carried out. The results show that the combination of AE and CNN has good performance with the increase of walking speed. In addition, a dimensionality reduction experiment based on AE is also carried out, and the optimal dimension is obtained.