Peng Zhang

Peng Zhang

Auto-generated profile, based on publicly available information

Peng Zhang

School of Mechanical Engineering, Tianjin University of Science and Technology, Tianjin, China

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

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

Representative Publications

View All Publications
27 January 2022 in IEEE Sensors Journal
Verified
Journal Article
Closed access
PZPeng Zhang
Peng Zhang
JZJunxia Zhang
Junxia Zhang
Ahmed Moneeb ElsabbaghAhmed Moneeb Elsabbagh
Ahmed Moneeb Elsabbagh
Surface Electromyography (sEMG) has been extensively used for gait analysis and robotic control. Since the low accuracy of intention recognition limits the development of exoskeleton robots, a dynamic adaptive neural network algorithm based on the multi-feature fusion of sEMG signals was proposed to realize the accurate identification of eight lower limb movements. Firstly, the original sEMG signals were collected from eighty volunteers through gait experiments. Secondly, the feature parameters of time domain, frequency domain, and sample entropy were extracted and merged. Then, the improved differential evolution (DE) algorithm was proposed to optimize the weight values of each feature for data fusion. Thirdly, the backpropagation neural network(BPNN) used in this study is the most common model for intention recognition. However, BPNN has the problems of low accuracy and slow convergence speed, this paper proposed a novel dynamic adaptive neural network model (GA-DANN). The novel model uses genetic algorithm (GA) to optimize the weights of the BPNN to accelerate convergence. The dynamic adaptive learning rate was introduced to overcome the difficulty of determining the learning rate. The experimental results show that the proposed algorithm with multi-feature fusion information could accurately distinguish eight lower limb motion intentions, and the averaged recognition accuracy reaches 94.89%, which is 10% higher than the traditional BPNN. The averaged recognition time is only 109.67ms. This research will provide technical support for the application of rehabilitation robots.

24 August 2020 in IEEE Access
Verified
Journal Article
Open access
PZPeng Zhang
Peng Zhang
JZJunxia Zhang
Junxia Zhang
ZZZunhao Zhang
Zunhao Zhang
In accordance with the movement coordination principle of both lower limbs, a complete radial basis functions neural network based adaptive sliding mode control strategy (RBFVSMC) is proposed. The movement information on the non-affected side of patients is detected to drive the rehabilitation training. The nonlinear mathematical model of the rehabilitation robot system is firstly described. Based on the robotic dynamic model, a variable sliding mode control (VSMC) is proposed to stabilize the system. To reduce the buffeting problem caused by VSMC, the universal approximation of RBFNN is used to approach and compensate external disturbances and uncertainties. Besides, the buffeting phenomenon of sliding mode control is alleviated by replacing the sign function with a saturation function. The final asymptotic stability is guaranteed with Lyapunov criteria. Compared to proportional-integral-derivative (PID), radial basis functions neural network (RBFNN), continuous terminal SMC (CNTSMC), and decentralized adaptive robust controller (NDOBCTC), the effectiveness of the overall control scheme is demonstrated by co-simulation and human experiment in accordance to track following performance and disturbances rejection ability.

21 November 2023 in IEEE Sensors Journal
Verified
Journal Article
Closed access
PZPeng Zhang
Peng Zhang
JZJunxia Zhang
Junxia Zhang
JJJun Jia
Jun Jia
Stair walking, as the most common gait pattern in daily life, also has a positive effect on the recovery of limb function in patients with lower limb disabilities. To realize steady walk on stairs, the most important problem is the structure design and gait planning of lower limb exoskeleton. A novel assisted exoskeleton system was proposed based on biomechanics characteristics. The four-point segmented gait generation strategy based on cubic spline interpolation was proposed, and the gait trajectory during stair walking was described. To ensure walking stability, the constraint relationship between the swing leg and the stair step was constructed. In addition, the constraint function was constructed for stable walking in horizontal and vertical directions based on Zero Moment Point (ZMP). Then, the multi-coupled continuous dynamic primitive gait learning strategy (PIRDMPs) was proposed to achieve fine-grained learning of stair gait trajectories. Finally, the proposed method was validated by human-machine wear experiments. Compared with the DMP method, the average trajectory fits for the hip and knee joints obtained based on the PIRDMPs method were increased by 11.28% and 7.17%, respectively. And the average deviation values of ZMP in the sagittal plane (ZMP-X) and coronal plane(ZMP-Y) of the man-machine system were reduced by 109.2 mm and 125.3 mm, respectively.