Neural Network Based Adaptive Actuator Fault Detection Algorithm for Robot Manipulators

Cho, Chang Nho; Hong, Ji; Kim, Hong Ju

doi:10.1007/s10846-018-0781-0

Cited by 37 publications

(19 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The resulting closed-loop overall system, after substituting of (37) (38) where q ∈ R n and z ∈ R n represent the link angles and elastic torque. The first simulation presented in Fig.…”

Section: A Design Examplementioning

confidence: 99%

“…3 shows the performance of the controls we have developed previously in the fault-free case. It is clear that the composite u comp asymptotically stabilizes the origin of the flexible-joint robotic system in (38) In order to verify the effectiveness of the control (37), two collision faults on the actuator F a , see Fig. 4(c) are injected.…”

Section: A Design Examplementioning

confidence: 99%

“…The first one is generated between the time instants 6 s and 8 s (before reaching the steady state) and the second one between the time instants 15 s and 18 s (after reaching the steady state): A collision defect is one of the common faults on which a robot collides with an object or a human, and it may leads to serious damage or injury. The generated parametric uncertainty is 10 % of the nominal value of the system parameters (38). Figure (4) shows that the composite control (37) becomes incapable of ensuring the convergence of the flexible-joint robotic system (38) in the presence of uncertainty and defect (see Fig.…”

Section: A Design Examplementioning

confidence: 99%

“…The generated parametric uncertainty is 10 % of the nominal value of the system parameters (38). Figure (4) shows that the composite control (37) becomes incapable of ensuring the convergence of the flexible-joint robotic system (38) in the presence of uncertainty and defect (see Fig. 4(a)).…”

Section: A Design Examplementioning

confidence: 99%

See 3 more Smart Citations

Reconfigurable control of flexible joint robot with actuator fault and uncertainty

Elghoul

Tellili

Abdelkrim

2019

Journal of Electrical Engineering

View full text Add to dashboard Cite

This paper presents the fault tolerant control (FTC) of a flexible joint robot using singular perturbation method in order to compensate for the lost performance due to the occurrence of actuator fault and the uncertainty. This FTC is based on Lyapunov redesign principle. The singular perturbation method is used to reduce the dynamic model of the flexible joint robot in a fast and slow subsystem. The time scale reduction of the flexible joint model is carried out when their joint stiffness is large enough and the singular perturbation parameter is set to zero. The fault-tolerant control structure in this paper is based on two parts. The first term described the composite control for the system without defect and without uncertainty which represents the sum between slow and fast controllers. While the second term of the fault tolerant command describes additive control designed to compensate for the fault effect of the actuator on the uncertain system. The additive approach is based on the Lyapunov theorem, which guarantees asymptotic stability despite the presence of actuator defects and the parametric uncertainty. The theoretical results are applied on a robot manipulator with a single flexible joint.

show abstract

Section: A Design Examplementioning

confidence: 99%

Section: A Design Examplementioning

confidence: 99%

Section: A Design Examplementioning

confidence: 99%

Section: A Design Examplementioning

confidence: 99%

See 2 more Smart Citations

Reconfigurable control of flexible joint robot with actuator fault and uncertainty

Elghoul

Tellili

Abdelkrim

2019

Journal of Electrical Engineering

View full text Add to dashboard Cite

show abstract

“…However, the fixed rules in FIS restrict the detection capability for unknown faults. To improve on this, neural network [11][12], support vector machine [13] and other machine learning tools are applied to train the model of fault detection. Data-based methods, which are essentially data-driven, analyze the observed data directly by using wavelet transform [14][15], auto regressive moving average [16] and so on.…”

Section: Introductionmentioning

confidence: 99%

A Novel Approach for Fault Detection in Integrated Navigation Systems

Zhu

Zhou

2020

IEEE Access

View full text Add to dashboard Cite

Detecting subsystem faults quickly is critical to the accuracy and reliability of integrated navigation systems. This paper, therefore, proposes an effective approach based on the novel test statistic to detect faults. Machine learning is introduced to estimate the innovation and its variance of local filter. The estimates combined with the actual ones are used to construct the test statistic, which is then proved to obey chi-square distribution. Thus fault detection can be realized by chi-square test. However, the special structure of the test statistic makes it sensitive to faults, even to the gradual faults. The experimental results demonstrate that the approach can detect faults quickly. Especially for gradual fault detection, the proposed test statistic has a marked superiority compared with the traditional test statistic of residual chi-square test.

show abstract

Actuator fault detection for masonry robot manipulator arm with the interval observer

Jinbao

Zhang

Hou

et al. 2022

Journal of Field Robotics

View full text Add to dashboard Cite

The main construction method of building wall is artificial masonry, the main problem is that the process is associated with low construction efficiency and poor safety, workers are prone fall from high altitude. The research of automatic masonry robot has become an urgent need. The masonry mechanical arm system is the main executing part of the masonry robot, special attention should be paid to the robot fault. Therefore, it is necessary to establish a suitable model to detect the actuator faults of the manipulator system. In this paper, a dynamic model of manipulator fault is presented and a fault detection scheme of masonry robot manipulator arm is proposed based on the model. The model is simplified by analyzing the state parameters of each joint during robot masonry and the interval observer with more design freedom was designed based on the established mathematical model of actuator faults. In this paper, a joint method for solving S and L matrices is proposed, which avoids the limitation of the traditional method for solving L matrices by twostep. In the presence of external interference, l 1 /H ∞ performance are introduced into the generation process of residual interval, and the interval observer has better disturbance robustness and fault sensitivity. Simulation experiments verify that the scheme can effectively detect the actuator fault of the manipulator, and experiments are carried out on a 6-axis manipulator. The experimental results show that when actuator faults occur at joints 2 and 3, the residual rapidly exceeds the threshold range, which proves the effectiveness of the fault detection scheme designed in this paper.

show abstract

Neural Network Based Adaptive Actuator Fault Detection Algorithm for Robot Manipulators

Cited by 37 publications

References 19 publications

Reconfigurable control of flexible joint robot with actuator fault and uncertainty

Reconfigurable control of flexible joint robot with actuator fault and uncertainty

A Novel Approach for Fault Detection in Integrated Navigation Systems

Actuator fault detection for masonry robot manipulator arm with the interval observer

Contact Info

Product

Resources

About