Clutch Mechanical Leg Neural Network Adaptive Robust Control of Shift Process for Driving Robot With Clutch Transmission Torque Compensation

Chen, Gang; Jiang, Junhao; Wang, Liangmo; Zhang, Weigong

doi:10.1109/tie.2021.3128909

Cited by 7 publications

(4 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…3 Property 2: The matrix _ D(q) À 2C(q, _ q) is skew symmetric, such that x T ( _ D(q) À 2C(q, _ q))x = 0, 8x, q, _ q 2 R n . 32 In this paper, it is supposed that the dynamics of the proposed compliant collaborative robot is uncertain, that is, the exact values of D(q), C(q, _ q), H and G(q) are unknown. The objective of this paper is to design torque t, as a result that, the proposed compliant collaborative robot can achieve better tracking performance under uncertainties.…”

Section: Dynamics Model Descriptionmentioning

confidence: 99%

“…Remark 2: The uncertain bounds are set simultaneously for the estimates of the weight matrices V D , V C , V H and V G along with errors r D (t), r C (t), r H (t) and r G (t), so that the estimated accuracy of D(q), C(q, _ q), H and G(q) is greatly improved. In the light of Stone Weierstrass theorem, 32 there are distinct weight matrices, which can well approximate different reference trajectories. This means that if the distinction among the estimate weight matrix and real weight matrix is in the uncertain bound of the weight matrix, and the uncertain bound of the approximation error is not violated, good control performance can be maintained.…”

Section: Controller Designmentioning

confidence: 99%

“…Property 2: The matrix true D . ( q ) − 2 C ( q , true q . ) is skew symmetric, such that x T ( true D . ( q ) − 2 C ( q , true q . ) ) x = 0 , ∀ x , q , true q . ∈ R n . 32…”

Section: Dynamics Model Descriptionmentioning

confidence: 99%

“…17 To improve the motion precision, an adaptive robust neural network control scheme was proposed for driving robot with torque compensation. 32 A robust tracking controller was proposed to make the tracking error bounded. 33 To improve tracking accuracy, an adaptive neural network controller was developed for the constrained robot manipulators.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Adaptive passive admittance control of a compliant collaborative robot

Xu,

2023

Proceedings of the Institution of Mechanical Engineers, Part C:

View full text Add to dashboard Cite

Compliance of the collaborative robots plays an important role in human-robot interaction. Due to the complex nonlinearities caused by the compliance, and parameter uncertainties, good performance control method for the compliant collaborative robot is always a challenging task. Therefore, to overcome the challenge of the compliance, and uncertain dynamics, this paper develops an adaptive passive admittance control (APAC) approach for a novel compliant collaborative robot, whose compliant mechanism is constituted of the inerters, springs, and dampers. First, through force-current analogy between mechanical network and electrical network, the compliance of the proposed collaborative robot with the inerters, springs, and dampers, is represented by a novel and simplified admittance matrix to facilitate dynamics model of the proposed collaborative robot. Then, for the proposed compliant collaborative robot, an APAC strategy is presented to dispose of the uncertain parameters of the dynamics. The inertial matrix, Coriolis matrix, admittance matrix, and gravity vector of the compliant collaborative robot are expressed as linear representation of orthogonal basis functions by using function approximation technique (FAT). The boundedness of the robot dynamics is proved by detailed stability analysis which uses Lyapunov function. Finally, simulation results indicate that the proposed APAC of the compliant collaborative robot is effective.

show abstract

Section: Dynamics Model Descriptionmentioning

confidence: 99%

Section: Controller Designmentioning

confidence: 99%

“…Property 2: The matrix true D . ( q ) − 2 C ( q , true q . ) is skew symmetric, such that x T ( true D . ( q ) − 2 C ( q , true q . ) ) x = 0 , ∀ x , q , true q . ∈ R n . 32…”

Section: Dynamics Model Descriptionmentioning

confidence: 99%