Deep GRU Neural-Network Prediction and Feedforward Compensation for Precision Multi-Axis Motion Control Systems

Hu, Chuxiong; Ou, Tiansheng; Chang, Haonan; Zhu, Yu; Zhu, Limin

doi:10.1109/tmech.2020.2975343

Cited by 69 publications

(25 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…4. First, we generate random NURBS geometric paths with random control points [22], which are composed of the independent variable vector t and the dependent variable vector y, described as follows:…”

Section: B Training Of Lstm-nnmentioning

confidence: 99%

“…1) different neural network features: We consider the proposed feature and others in the literature, including: i) reference velocities at the current instant and at the previous instants [28]; ii) reference position, velocity, and acceleration at the current instant [22]; iii) reference velocities at the current instant, at the previous and at the next instants, and the non-linear features (proposed). The comparative results for a butterfly curve are given in Fig.…”

Section: B Comparison Of Different Neural Network Features and Differ...mentioning

confidence: 99%

“…In [21], a feedforward input generation scheme based on neural network prediction was proposed for six-degree-of-freedom industrial robots, which was shown to have superior performance in position tracking and residual vibration suppression. In [22], a deep gated recurrent unit (GRU) neural network was adopted to predict the contour error of the multi-axis motion system, and the predicted contour error was used as pre-compensation. The neural network feature selection in these studies mainly adopts a heuristic approach and only considers linear features, such as velocity or acceleration, leading to low prediction accuracy.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Prediction and Compensation of Contour Error of CNC Systems Based on LSTM Neural-Network

et al. 2022

IEEE/ASME Trans. Mechatron.

View full text Add to dashboard Cite

show abstract

Section: B Training Of Lstm-nnmentioning

confidence: 99%

Section: B Comparison Of Different Neural Network Features and Differ...mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Prediction and Compensation of Contour Error of CNC Systems Based on LSTM Neural-Network

et al. 2022

IEEE/ASME Trans. Mechatron.

View full text Add to dashboard Cite

show abstract

“…We generate the random NURBS trajectory by using random control points [41], which are composed of independent variable vector and dependent variable vector, and the independent variable vector t is as follows…”

Section: Training and Testing Datamentioning

confidence: 99%

Reference modification for trajectory tracking using hybrid offline and online neural networks learning

Huang

Zhong

et al. 2022

Neural Comput & Applic

View full text Add to dashboard Cite

In this paper, we propose a hybrid offline/online neural networks learning method, which combines complementary advantages of two types of neural networks (NNs): deep NN (DNN) and single-layer radial basis function NN (RBFNN). Firstly, after analyzing the mechatronic system’s model, we select reasonable features as the input of the DNN to learn the inverse dynamic characteristics of the closed-loop system offline, so as to establish the mapping between the desired trajectory and the reference trajectory of the system. The trained DNN is used to generate a new reference trajectory and compensate for the tracking error in advance, which can speed up the convergence of online learning control based on RBFNN. This reference trajectory is further modified iteratively when the tracking task is repeated. For this purpose, a single-layer RBFNN model is established, and an online learning algorithm is developed to update the RBFNN parameters. The proposed hybrid offline/online NN method can improve the tracking performance of mechatronic systems by modifying the reference trajectory on top of the baseline controller without affecting the system stability. To verify the effectiveness of this method, we conduct experiments on a piezoelectric drive platform.

show abstract

“…Further, the backlash, friction, deformation and other nonlinear factors in the surface of the mechanical transmission structure can result in an unexpected shortterm standstill phenomenon when doing a speed reversal movement [20]. The short-term standstill phenomenon means lost motion, which is impossible to be eliminated by choosing a position controller due to the nonlinear factors [21], [22]. The lost motion will introduce a typical contour error-over-quadrant error in the load side.…”

Section: Introductionmentioning

confidence: 99%

Vibration Suppression and Over-Quadrant Error Mitigation Methods for a Ball-Screw Driven Servo System With Dual-Position Feedback

Yang

Liu

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Deep GRU Neural-Network Prediction and Feedforward Compensation for Precision Multi-Axis Motion Control Systems

Cited by 69 publications

References 37 publications

Prediction and Compensation of Contour Error of CNC Systems Based on LSTM Neural-Network

Prediction and Compensation of Contour Error of CNC Systems Based on LSTM Neural-Network

Reference modification for trajectory tracking using hybrid offline and online neural networks learning

Vibration Suppression and Over-Quadrant Error Mitigation Methods for a Ball-Screw Driven Servo System With Dual-Position Feedback

Contact Info

Product

Resources

About