An indirect adaptive neural control of a visual-based quadrotor robot for pursuing a moving target

Shirzadeh, Masoud; Amirkhani, Abdollah; Jalali, Aliakbar; Mosavi, Mohammad Reza

doi:10.1016/j.isatra.2015.10.011

Cited by 46 publications

(15 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another way for neural adapting is based on the Jacobian information, the sensitivity of the system output to each control input ∂y/∂u. Shirzadeh et al [61] proposed a control for a quadrotor type aerial robot that is in pursuit of a mov-ing target. One feed-forward model is used for identification of the Jacobian information and the second model is designed for the generation of the control signal.…”

Section: Neural Feedback Controlmentioning

confidence: 99%

A review of adaptive neural control applied to proton exchange membrane fuel cell systems

Lin-Kwong-Chon

Grondin-Pérez

Kadjo

et al. 2019

Annual Reviews in Control

View full text Add to dashboard Cite

Section: Neural Feedback Controlmentioning

confidence: 99%

A review of adaptive neural control applied to proton exchange membrane fuel cell systems

Lin-Kwong-Chon

Grondin-Pérez

Kadjo

et al. 2019

Annual Reviews in Control

View full text Add to dashboard Cite

“…where d , together with P d is selected as the designated trajectory, whereas the desired angles d and d , being treated as the control inputs for translational subsystem, can be solved by equations (16) and (18), where 4À6 will be given as equation 43in subsequent analysis. Therefore, the three equations (10) to (12) contains only three unknown factors, Figure 2.…”

Section: Control Schemementioning

confidence: 99%

“…9 adopted an extended state observer to estimate the state and the aerodynamic disturbances, which is considered in both the attitude and trajectory control. Others, such as H∞, 10,11 feedback control, 12,13 intelligent control, 14–16 and hybrid strategies 15–22 have been attempted for the flight robust control of quadrotor to improve the performance. Nevertheless, more attentions should be devoted to the smooth control inputs generated by controller, which is essential for the stability and reliability of the controlled system.…”

Section: Introductionmentioning

confidence: 99%

Binary observers based control for quadrotor unmanned aerial vehicle with disturbances and measurement delay

Zhang

Zhou

et al. 2017

Proceedings of the Institution of Mechanical Engineers, Part G:

View full text Add to dashboard Cite

This paper presents a binary observers based linearization feedback sliding mode controller for quadrotor subjected to disturbances and measurement delay. The quadrotor unmanned aerial vehicle is divided into two subsystems, i.e. rotational subsystem and translational subsystem. Disturbance observer based controller is proposed for the former sensitive to disturbances, and binary observers based one for the latter subjected to both disturbances and measurement delay. Binary observers, consisting of disturbance observer and delayed output observer with their interactions being considered, promise exponential convergences of estimations on disturbance and output to their real values, which are compensated in the feedback control loop. The stabilities of the closed-loop subsystems are demonstrated based on Lyapunov theory. In the presence of both disturbances and measurement delay, simulations are carried out to corroborate the effectiveness and robustness of the proposed controller with superior control inputs.

show abstract

“…However, the above-mentioned control synchronization study on the exciters is only in the linear vibration system, the control synchronization study of the exciters in the nonlinear vibration system have barely been studied. The sliding mode control combined with radial basis neural network approximation can suppress the system's chattering and make the system's response more stable, which is very suitable for control of nonlinear systems [34]- [36]. Motivated by the above observations, this paper establishes an RBFN-AGSMA controller for the synchronization problem of two homodromy exciters in nonlinear vibration systems.…”

Section: Introductionmentioning

confidence: 95%

Control Synchronization of Two Nonidentical Homodromy Exciters in Nonlinear Coupled Vibration System

et al. 2019

View full text Add to dashboard Cite

In order to analyze synchronization control problems of two non-identical homodromy eccentric rotors (ERs) in a nonlinearly coupled system of vibrating machinery-part (NCS-VMP), a new electromechanical coupling nonlinear dynamic model considering nonlinear acting force of the part and nonlinear support is established, NCS-VMP's complex control is converted into rotating speed and phase synchronous control of two homodromy non-identical exciters. By considering the dynamic interactions among the vibration body, the part and the ERs, the nonlinear dynamical equation of the NCS-VMP are established. An accurate synchronization control of speed and phase method are proposed for two homodromy ERs in NCS-VMP. The precise speed and phase synchronization control is mainly reflected as: The cross-coupling control strategy is used which considering the coupling effect between two co-rotating exciters. The radial basis function network adaptive global sliding mode algorithm (RBFN-AGSMA) is used to adaptively approximate the total uncertainty of system including the nonlinear support and the nonlinear force of the parts, which can effectively reduce the estimation error. The radial basis function network method can suppress the jitter of the system and make the influence of the system more stable after replacing the sign function. The stability of RBFN-AGSMA controller is proved by Lyapunov theory. The controller's property is verified through numerical methods and taking the sliding mode control (SMC) algorithm into comparison. Results indicate that the designed control method can reduce chattering clearly compared with the SMC algorithm, and it is capable to improve the control accuracy of two non-identical homodromy exciters. By studying the effects of parameter change in the NCS-VMP on the system, the strong robustness of RBF network global sliding mode controller to parameter perturbations is proved. It is proposed that RBFN-AGSMA controller can control two nonidentical homodromy ERs in NCS-VMP to achieve accurate vibration trajectory in the working direction.

show abstract

An indirect adaptive neural control of a visual-based quadrotor robot for pursuing a moving target

Cited by 46 publications

References 31 publications

A review of adaptive neural control applied to proton exchange membrane fuel cell systems

A review of adaptive neural control applied to proton exchange membrane fuel cell systems

Binary observers based control for quadrotor unmanned aerial vehicle with disturbances and measurement delay

Control Synchronization of Two Nonidentical Homodromy Exciters in Nonlinear Coupled Vibration System

Contact Info

Product

Resources

About