Adaptive Neural Impedance Control of a Robotic Manipulator With Input Saturation

He, Wei; Dong, Yiting; Sun, Changyin

doi:10.1109/tsmc.2015.2429555

Cited by 721 publications

(290 citation statements)

References 60 publications

Supporting

Mentioning

280

Contrasting

Order By: Relevance

“…In Equation (2), the measurement process is described by a non-linear and time-dependent function (h k ), and the measurement noise vector is represented by n k . In a Bayesian estimation, state vector at time k is estimated from all the measurements (z 1:k ) up to and including time k. At each point in time, a probability distribution over x k , called belief, bel(x k ) = p(x k |z 1:k ), represents the uncertainty.…”

Section: Bayes Filtermentioning

confidence: 99%

“…For this simulation we used 4 landmarks at (0, 0), (500, 0), (0, 500), (500, 500) in a virtual workspace of 500 cm 2 . Noise values in the range of (0-30) cm were added in the distance of the Wheeled Mobile Robot (WMR) from the nearest landmark in accordance with the observed noise of Real-Time Location System (RTLS) discussed above.…”

Section: Distance Noise Effect On Pf Performancementioning

confidence: 99%

“…The loss function is mse. X r , Y r is a random point from a 30 cm 2 region, which represents the position of the Wheeled Mobile Robot (WMR), X m , Y m is current weighted mean of the PF, while [0][1][2][3][4][5][6][7] is the output of the DQN. γ is 0.95, exploration-rate is 1.0, exploration-minimum is 0.01, exploration-decay is 0.995, sample batch size is 100 and number of episodes are 10,000.…”

Section: Dqn Parametersmentioning

confidence: 99%

“…Wheeled Mobile Robot (WMR) has a broad spectrum of applications in both civilian and military engineering areas, such as search and rescue, mine clearance, scout, superstores, restaurants, and space exploration [1,2]. Wheel slip becomes a problem when the pure rolling conditions are not satisfied.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Robust Localization, Slip Estimation, and Compensation System for WMR in the Indoor Environments

Ullah

Zhang

et al. 2018

Symmetry

View full text Add to dashboard Cite

Abstract:A novel approach is proposed for the path tracking of a Wheeled Mobile Robot (WMR) in the presence of an unknown lateral slip. Much of the existing work has assumed pure rolling conditions between the wheel and ground. Under the pure rolling conditions, the wheels of a WMR are supposed to roll without slipping. Complex wheel-ground interactions, acceleration and steering system noise are the factors which cause WMR wheel slip. A basic research problem in this context is localization and slip estimation of WMR from a stream of noisy sensors data when the robot is moving on a slippery surface, or moving at a high speed. DecaWave based ranging system and Particle Filter (PF) are good candidates to estimate the location of WMR indoors and outdoors. Unfortunately, wheel-slip of WMR limits the ultimate performance that can be achieved by real-world implementation of the PF, because location estimation systems typically partially rely on the robot heading. A small error in the WMR heading leads to a large error in location estimation of the PF because of its cumulative nature. In order to enhance the tracking and localization performance of the PF in the environments where the main reason for an error in the PF location estimation is angular noise, two methods were used for heading estimation of the WMR (1): Reinforcement Learning (RL) and (2): Location-based Heading Estimation (LHE). Trilateration is applied to DecaWave based ranging system for calculating the probable location of WMR, this noisy location along with PF current mean is used to estimate the WMR heading by using the above two methods. Beside the WMR location calculation, DecaWave based ranging system is also used to update the PF weights. The localization and tracking performance of the PF is significantly improved through incorporating heading error in localization by applying RL and LHE. Desired trajectory information is then used to develop an algorithm for extracting the lateral slip along X-and Y-axis from the PF estimated position of the WMR, the lateral slip along X-and Y-axis is then used to take some corrective measures. Lateral slip information is also used to find the direction along which WMR has to move to get back along the desired trajectory. Simulation results show that our proposed LHE and RL heading estimation methods significantly improve the PF localization and tracking performance on a slippery surface in both indoor and outdoor environments. The simulation results also show that the accurate locations of WMR and desired path information are used to estimate and compensate the lateral slip.

show abstract

Section: Bayes Filtermentioning

confidence: 99%

Section: Distance Noise Effect On Pf Performancementioning

confidence: 99%

Section: Dqn Parametersmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Robust Localization, Slip Estimation, and Compensation System for WMR in the Indoor Environments

Ullah

Zhang

et al. 2018

Symmetry

View full text Add to dashboard Cite

show abstract

“…However, serious nonlinear behavior, such as control input saturation [1], state constraint, valve opening, nonlinear friction [2], and model uncertainty [3] (load changes, the parameters variation and the element parameter uncertainty [4] caused by the abrasive, containing external disturbance [5,6], leakage, and other uncertain nonlinear elements), restricts the development of high-performance closed loop system controller [7,8].…”

Section: Introductionmentioning

confidence: 99%

Adaptive Neural Network Control for Nonlinear Hydraulic Servo-System with Time-Varying State Constraints

2017

Complexity

View full text Add to dashboard Cite

An adaptive neural network control problem is addressed for a class of nonlinear hydraulic servo-systems with time-varying state constraints. In view of the low precision problem of the traditional hydraulic servo-system which is caused by the tracking errors surpassing appropriate bound, the previous works have shown that the constraint for the system is a good way to solve the low precision problem. Meanwhile, compared with constant constraints, the time-varying state constraints are more general in the actual systems. Therefore, when the states of the system are forced to obey bounded time-varying constraint conditions, the high precision tracking performance of the system can be easily realized. In order to achieve this goal, the time-varying barrier Lyapunov function (TVBLF) is used to prevent the states from violating time-varying constraints. By the backstepping design, the adaptive controller will be obtained. A radial basis function neural network (RBFNN) is used to estimate the uncertainties. Based on analyzing the stability of the hydraulic servo-system, we show that the error signals are bounded in the compacts sets; the time-varying state constrains are never violated and all singles of the hydraulic servo-system are bounded. The simulation and experimental results show that the tracking accuracy of system is improved and the controller has fast tracking ability and strong robustness.

show abstract

References

2019

Kinematic Control of Redundant Robot Arms Using Neural Networks

View full text Add to dashboard Cite

Adaptive Neural Impedance Control of a Robotic Manipulator With Input Saturation

Cited by 721 publications

References 60 publications

A Robust Localization, Slip Estimation, and Compensation System for WMR in the Indoor Environments

A Robust Localization, Slip Estimation, and Compensation System for WMR in the Indoor Environments

Adaptive Neural Network Control for Nonlinear Hydraulic Servo-System with Time-Varying State Constraints

References

Contact Info

Product

Resources

About