Deep Convolutional Identifier for Dynamic Modeling and Adaptive Control of Unmanned Helicopter

Kang, Yu; Chen, Shaofeng; Wang, Xuefeng; Cao, Yang

doi:10.1109/tnnls.2018.2844173

Cited by 43 publications

(14 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The neural network is a universal function approximator with a sequential model structure [30], which is described as follows Within the area of control, the use of the convolutional neural networks (CNNs) has had a very strong impact on tasks such as system identification [31], path generation [32] and so on. The biggest difference between CNN and general neural networks lies in the use of the convolution layer, which is mainly used to process two-dimensional data, such as images and time sequence data [33].…”

Section: Deep Neural Networkmentioning

confidence: 99%

Deep representation-based packetized predictive compensation for networked nonlinear systems

Chen

Cao

Kang

et al. 2020

Neural Comput & Applic

Self Cite

View full text Add to dashboard Cite

The design of networked nonlinear control system is a very challenging problem due to the coupling of the system uncertainties (e.g., model accuracy, noise, nonlinearity) and network effects (e.g., packet dropouts, time delay). In this paper, a deep representation-based predictive compensation method is proposed for networked nonlinear systems with packet-dropouts in the feedback and forward channel. Different from the existing compensation methods based on openloop prediction, the proposed method is based on feedback compensation and does not require a nominal system model, so that it can avoid the coupling of system uncertainties and network effects as well as the occupation of network bandwidth. Specifically, a deep sequence to sequence learning scheme is firstly employed to encode the correlations of state and control sequences into deep feature representations. Furthermore, within the embedding space spanned by the learned features, according to the state-control sequence of each sampling step, a prediction of the next control command is generated as compensation for packet dropout. The stability of the overall system is rigorously proved by the Lyapunov theory, which reveals that the control errors for the networked control systems with packet dropouts asymptotic converge to a small neighborhood of the origin. We further evaluate the performance of the proposed strategy on a wheeled mobile robots simulation platform, and the experimental results demonstrate that our method can achieve high compensation accuracy and robustness concerning packet dropouts, even in the case of the maximum continuous packet dropouts specified by the network communication protocol.

show abstract

Section: Deep Neural Networkmentioning

confidence: 99%

Deep representation-based packetized predictive compensation for networked nonlinear systems

Chen

Cao

Kang

et al. 2020

Neural Comput & Applic

Self Cite

View full text Add to dashboard Cite

show abstract

“…In our previous work [ 24 ], a modeling scheme for nonlinear systems is proposed using CNN with real values. In [ 25 ], CNN is trained to model uncertainties in the dynamic system. These uncertainties are over passed with the CNN because of the properties of share weights and sparse connectivity.…”

Section: Introductionmentioning

confidence: 99%

Complex Valued Deep Neural Networks for Nonlinear System Modeling

López-Pacheco

2021

Neural Process Lett

View full text Add to dashboard Cite

Deep learning models, such as convolutional neural networks (CNN), have been successfully applied in pattern recognition and system identification recent years. But for the cases of missing data and big noises, CNN does not work well for dynamic system modeling. In this paper, complex valued convolution neural network (CVCNN) is presented for modeling nonlinear systems with large uncertainties. Novel training methods are proposed for CVCNN. Comparisons with other classical neural networks are made to show the advantages of the proposed methods.

show abstract

“…12 As an alternative, different types of deep neural networks, such as deep convolutional neural networks, have been utilized to identify complicated system dynamics. 13 Nevertheless, the training process of a single neural network (such as a deep neural network) for identification of a complex dynamic system may also converge to local minima due to the difficulty of finding the proper network structure in each application. Furthermore, as the neural networks result in a black-box identification scheme, there is no direct relationship between a specific part of the system and the network parameters.…”

Section: Introductionmentioning

confidence: 99%

A self-organizing multi-model ensemble for identification of nonlinear time-varying dynamics of aerial vehicles

Emami

Ahmadi

2020

Proceedings of the Institution of Mechanical Engineers, Part I:

View full text Add to dashboard Cite

This article presents a novel identification approach which can deal with nonlinear and time-varying characteristics of complex dynamic systems, especially an aerial vehicle in the entire flight envelope. A set of local sub-models are first developed at different operating points of the system, and subsequently a self-organizing multi-model ensemble is introduced to aggregate the outputs of the local models as a single model. The number of employed local models in the proposed multi-model ensemble is optimized using a novel self-organizing approach. Also, wavelet neural networks, which combine both the universal approximation property of neural networks and the wavelet decomposition capability, are used as the local models of the proposed method. In addition, a generalized online sequential extreme learning machine is adopted in the introduced approach to determine the optimal validity function of the local models at each time step. Finally, the introduced self-organizing multi-model ensemble is applied to the NASA Generic Transport Model as a complex nonlinear system to demonstrate the effectiveness of the proposed identification approach. Furthermore, the results obtained from the conventional artificial neural networks are carefully compared with those from the wavelet neural networks, which are employed as the local models of the introduced multi-model ensemble. The simulation results suggest that the introduced wavelet neural network–based self-organizing multi-model ensemble can be used satisfactorily as the prediction model of model-based control systems for long prediction horizons.

show abstract

Deep Convolutional Identifier for Dynamic Modeling and Adaptive Control of Unmanned Helicopter

Cited by 43 publications

References 29 publications

Deep representation-based packetized predictive compensation for networked nonlinear systems

Deep representation-based packetized predictive compensation for networked nonlinear systems

Complex Valued Deep Neural Networks for Nonlinear System Modeling

A self-organizing multi-model ensemble for identification of nonlinear time-varying dynamics of aerial vehicles

Contact Info

Product

Resources

About