Swing Phase Control of Semi-Active Prosthetic Knee Using Neural Network Predictive Control With Particle Swarm Optimization

Ekkachai, Kittipong; Nilkhamhang, Itthisek

doi:10.1109/tnsre.2016.2521686

Cited by 36 publications

(31 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To capture these behaviors of MR damper, the elementary hysteresis model (EHM) based feed-forward neural network (FNN) model is used in our simulation. It was proposed in Ekkachai et al (2012) and modified in Ekkachai and Nilkhamhang (2016). The model consists of two FNNs.…”

Section: System Descriptionmentioning

confidence: 99%

“…The gait data used in this study are also normal gait data collected from Ekkachai and Nilkhamhang (2016) for convenience in comparison study of the controller. In this manner, the proposed controller performance can be compared to the previous method with same dataset.…”

Section: Datasetmentioning

confidence: 99%

“…Several studies have tried to apply machine learning algorithm to control prosthetic (Ekkachai and Nilkhamhang, 2016;Wen et al, 2017Wen et al, , 2019. Neural network predictive control (NNPC) was employed as a control structure for the swing phase in the prosthetic knee (Ekkachai and Nilkhamhang, 2016). The swing phase model was constructed following a feed-forward neural network structure in which the input and the output were the knee angle, control voltage, and prediction of future knee angle.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Reinforcement Q-Learning Control With Reward Shaping Function for Swing Phase Control in a Semi-active Prosthetic Knee

et al. 2020

Self Cite

View full text Add to dashboard Cite

In this study, we investigated a control algorithm for a semi-active prosthetic knee based on reinforcement learning (RL). Model-free reinforcement Q-learning control with a reward shaping function was proposed as the voltage controller of a magnetorheological damper based on the prosthetic knee. The reward function was designed as a function of the performance index that accounts for the trajectory of the subject-specific knee angle. We compared our proposed reward function to a conventional single reward function under the same random initialization of a Q-matrix. We trained this control algorithm to adapt to several walking speed datasets under one control policy and subsequently compared its performance with that of other control algorithms. The results showed that our proposed reward function performed better than the conventional single reward function in terms of the normalized root mean squared error and also showed a faster convergence trend. Furthermore, our control strategy converged within our desired performance index and could adapt to several walking speeds. Our proposed control structure has also an overall better performance compared to user-adaptive control, while some of its walking speeds performed better than the neural network predictive control from existing studies.

show abstract

Section: System Descriptionmentioning

confidence: 99%

Section: Datasetmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Reinforcement Q-Learning Control With Reward Shaping Function for Swing Phase Control in a Semi-active Prosthetic Knee

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…For example, M. Islam et al used velocity and segmentation angle of the ankle as inputs of the neural network, which can automatically detect the gait mode of the powered ankle-foot orthosis [22]. K. Ekkachai et al proposed a neural network predictive control of the MR damper prosthetic knee with inputs of the knee angle and control voltage, which can provide amputees with better knee angle trajectories than conventional open-loop controllers [23]. Apart from using only kinematic data, dynamics data such as joint forces can also be used for the neural network control.…”

Section: Introductionmentioning

confidence: 99%

Design and Speed-Adaptive Control of a Powered Geared Five-Bar Prosthetic Knee Using BP Neural Network Gait Recognition

Sun

Huang

Jia

et al. 2019

Sensors

View full text Add to dashboard Cite

To improve the multi-speed adaptability of the powered prosthetic knee, this paper presented a speed-adaptive neural network control based on a powered geared five-bar (GFB) prosthetic knee. The GFB prosthetic knee is actuated via a cylindrical cam-based nonlinear series elastic actuator that can provide the desired actuation for level-ground walking, and its attitude measurement is realized by two inertial sensors and one load cell on the prosthetic knee. To improve the performance of the control system, the motor control and the attitude measurement of the GFB prosthetic knee are run in parallel. The BP neural network uses input data from only the GFB prosthetic knee, and is trained by natural and artificially modified various gait patterns of different able-bodied subjects. To realize the speed-adaptive control, the prosthetic knee speed and gait cycle percentage are identified by the Gaussian mixture model-based gait classifier. Specific knee motion control instructions are generated by matching the neural network predicted gait percentage with the ideal walking gait. Habitual and variable speed level-ground walking experiments are conducted via an able-bodied subject, and the experimental results show that the neural network control system can handle both self-selected walking and variable speed walking with high adaptability.

show abstract

“…Because neural network can fully approximate complex nonlinear systems, it has the characteristics of learning and adapting to the dynamic characteristics of uncertain systems, strong robustness and fault tolerance. So it has become a powerful tool for predictive control [21]- [23]. Some literatures choose BP neural network as the predictive model [24].…”

Section: Introductionmentioning

confidence: 99%

Multi-Step Predictive Current Control for NPC Grid-Connected Inverter

Wang

2019

IEEE Access

View full text Add to dashboard Cite

Aiming at improving the quality of output current when the inverter is connected to the grid, this paper proposes a control strategy of neutral point clamped (NPC) grid-connected inverter based on radial basis function neural network (RBFNN) multi-step predictive control. Firstly, the predictive control model for output current of NPC inverter is constructed by αβ coordinate transformation. Then the future values of output current are predicted for each voltage vectors, using the RBFNN prediction model. After the predicted values are obtained, a cost function f is calculated for each voltage vectors and the optimal voltage vector which minimizes cost function is selected. The switch state corresponding to the optimal voltage vector is applied to the inverter in the next sampling period. Thereby the control of grid-connected output current can be achieved. The simulation results show that the RBFNN multi-step predictive control method improves the capability of current tracking and reduces the harmonic rate of output current. The proposed method can balance output current under the condition of grid voltage imbalance. INDEX TERMS Inverter, RBF neural network, predictive control, optimal voltage vector, three-phase current balance.

show abstract

Swing Phase Control of Semi-Active Prosthetic Knee Using Neural Network Predictive Control With Particle Swarm Optimization

Cited by 36 publications

References 21 publications

Reinforcement Q-Learning Control With Reward Shaping Function for Swing Phase Control in a Semi-active Prosthetic Knee

Reinforcement Q-Learning Control With Reward Shaping Function for Swing Phase Control in a Semi-active Prosthetic Knee

Design and Speed-Adaptive Control of a Powered Geared Five-Bar Prosthetic Knee Using BP Neural Network Gait Recognition

Multi-Step Predictive Current Control for NPC Grid-Connected Inverter

Contact Info

Product

Resources

About