Adaptive compensation of modeled friction using a RBF neural network approximation

Vitiello, Valentina; Tornambè, Antonio

doi:10.1109/cdc.2007.4434037

Cited by 9 publications

(4 citation statements)

References 23 publications

(34 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The predicted torque is calculated using Equation ( 5). When it comes to the test set, a different excitation trajectory can be generated by adding a new constraint based on Equation (6). To apply a new constraint q i (t = 3.8) = 0, the test set trajectory is shown in Figure 4.…”

Section: Excitation Trajectory Generationmentioning

confidence: 99%

“…Many researchers have proposed different error compensation policies [4,5]. Machine learning methods, such as back propagation (BP) neural network [6] and radial basis function (RBF) network [7,8], are also used to fix or compensate errors. C Li et al [9] present a novel dynamic modeling method by using a recurrent neural network (RNN) with an incomplete state system variables observation.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Systematic Error Compensation Strategy Based on an Optimized Recurrent Neural Network for Collaborative Robot Dynamics

Zhang

Hou

et al. 2020

Applied Sciences

View full text Add to dashboard Cite

Robot dynamics and its parameter identification are of great significance to the realization of optimal control and human–machine interaction. The objective of this research is to address the shortcomings of establishing and identifying the self-developed six-degree-of-freedom (6-DoF) collaborative robot dynamics, which leads to a large error in the predicted torque of the proposed robot. A long short-term memory (LSTM) in an optimized recurrent neural network (RNN) is proposed to compensate the dynamic model of the proposed 6-DoF collaborative robot based on the consideration of gravity, Coriolis force, inertial force, and friction force. The analysis and experimental findings provide promising results. The compensated collaborative robot dynamic model based on LSTM in an optimized RNN displays a good prediction on the actual torque, and the root-mean-square (RMS) error between predicted and actual torques are reduced by 61.8% to 78.9% compared to the traditional dynamic model. Results of the experimental applications demonstrate the validity of the proposed systematic error compensation strategy.

show abstract

Section: Excitation Trajectory Generationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A Systematic Error Compensation Strategy Based on an Optimized Recurrent Neural Network for Collaborative Robot Dynamics

Zhang

Hou

et al. 2020

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…Using this property, NNs are used to build compensators for the friction models. [42][43][44] NNs are also used to handle the unknown dynamics including friction discontinuity. 30 However, the approximation error exists and depends on the structure of the NN.…”

Section: Related Workmentioning

confidence: 99%

Joint friction estimation for walking bipeds

Hashlamon¹,

Erbatur²

2014

Robotica

View full text Add to dashboard Cite

SUMMARYThis paper proposed a new approach for the joint friction estimation of non-slipping walking biped robots. The proposed approach is based on the combination of a measurement-based strategy and a model-based method. The former is used to estimate the joint friction online when the foot is in contact with the ground, while the latter adopts a friction model to represent the joint friction when the leg is swinging. The measurement-based strategy utilizes the measured ground reaction forces (GRF) and the readings of an inertial measurement unit (IMU) located at the robot body. Based on these measurements, the joint angular accelerations and the body attitude and velocity are estimated. The aforementioned measurements and estimates are used in a reduced dynamical model of the biped. However, when the leg is swinging, this strategy is inapplicable. Therefore, a friction model is adopted. Its parameters are identified adaptively using the estimated online friction whenever the foot is in contact. The estimated joint friction is used in the feedback torque control signal. The proposed approach is validated using the full-dynamics of 12-DOF biped model. By using this approach, the robot center of mass (CoM) position error is reduced by 10% which demonstrates the effectiveness of this approach.

show abstract

“…Neural networks have the ability to approximate nonlinear function. As in [5], RBF network was used to approximate the parameters for nonlinear characteristics of friction. Therefore, RBF network is reasonably suitable to be used in nonlinear switching law design.…”

Section: Smc Control Law Designmentioning

confidence: 99%

Study on Sliding Mode Control with RBF Network for DSTATCOM

Liu

Chen

et al. 2012

AMR

View full text Add to dashboard Cite

For the problems of parameters disturbance, nonlinear and uncertainty of distribution static compensator (DSTATCOM), this paper studies on sliding mode control based on radial basis function (RBF) network. The fast tracking of DSTATCOM reactive current is achieved by using of RBF neural networks and equivalent sliding mode control. The method has a strong adaptability and robustness for load disturbances and system parameters change, and integrates the advantages of neural network and sliding mode control, so it is an ideal intelligent control strategy .The MATLAB simulation results show that the controller has good dynamic and static quality, and provides an effective way for improving the performance of DSTATCOM.

show abstract

Adaptive compensation of modeled friction using a RBF neural network approximation

Cited by 9 publications

References 23 publications

A Systematic Error Compensation Strategy Based on an Optimized Recurrent Neural Network for Collaborative Robot Dynamics

A Systematic Error Compensation Strategy Based on an Optimized Recurrent Neural Network for Collaborative Robot Dynamics

Joint friction estimation for walking bipeds

Study on Sliding Mode Control with RBF Network for DSTATCOM

Contact Info

Product

Resources

About