16,472 Hits in 6.7 sec

Memory Augmented Neural Network Adaptive Controllers: Performance and Stability [article]

Deepan Muthirayan, Pramod P. Khargonekar
2021 arXiv   pre-print
The proposed architecture, in the setting of standard Neural Network (NN) based adaptive control, augments an external working memory to the NN.  ...  Through extensive simulations and specific metrics, such as peak deviation and settling time, we show that memory augmentation improves learning significantly.  ...  In Section II we propose the Memory Augmented Neural Network (MANN) adaptive controller.  ... 
arXiv:1905.02832v16 fatcat:cklxss6wsrfwbaqb4yqwl2sjba

Memory Augmented Neural Network Adaptive Controller for Strict Feedback Nonlinear Systems [article]

Deepan Muthirayan, Pramod P. Khargonekar
2020 arXiv   pre-print
Here, each NN, in the backstepping NN adaptive controller, is augmented with an external working memory.  ...  We propose a novel backstepping memory augmented NN (MANN) adaptive control method for the control of strict feedback non-linear systems.  ...  In a very recent paper [25] , we introduced a memory augmented neural network adaptive controller for model reference adaptive control (MRAC) and robot arm trajectory tracking controller.  ... 
arXiv:1906.05421v7 fatcat:dfrpysrf7rhgdoaqpemhtzkclm

Improved Attention Models for Memory Augmented Neural Network Adaptive Controllers [article]

Deepan Muthirayan, Scott Nivison, Pramod P. Khargonekar
2020 arXiv   pre-print
We introduced a working memory augmented adaptive controller in our recent work. The controller uses attention to read from and write to the working memory.  ...  Attention allows the controller to read specific information that is relevant and update its working memory with information based on its relevance.  ...  In Section II, we revisit the Memory Augmented Neural Network (MANN) adaptive controller proposed in our recent works [2] , [3] .  ... 
arXiv:1910.01189v7 fatcat:6xjsmsxbbjfgbmygkebccozu6y

Memory Augmented Neural Network-Based Intelligent Adaptive Fault Tolerant Control for a Class of Launch Vehicles Using Second-Order Disturbance Observer

Haipeng Chen, Kang Chen, Wenxing Fu, Jian Zhang
2021 Mathematical Problems in Engineering  
Simulation results demonstrate the desired performance and the advantages of the proposed control algorithm.  ...  This paper focuses on the MANN-based intelligent adaptive fault tolerant control for a class of launch vehicles.  ...  Memory Augmented Neural Networks (MANNs).  ... 
doi:10.1155/2021/9961278 fatcat:vaaf4jhktvdjflq2n7yo74rnty

Adaptive output feedback control of nonlinear systems using neural networks

A. Calise, N. Hovakimyan, Hungu Lee
2000 Proceedings of the 2000 American Control Conference. ACC (IEEE Cat. No.00CH36334)  
To handle the uncertainty, the controller, in the form of a dynamic compensator, is augmented by a single hidden layer (SHL) neural network that adjusts on-line for unknown nonlinearities.  ...  The parameter update laws for a SHL neural network are derived from stability analysis. Simulations illustrate the theoretical results.  ...  Fig.4 represents the closed-loop performance of the output feedback controller without neural network. The tracking performance of the linear controller is very poor in this case.  ... 
doi:10.1109/acc.2000.879146 fatcat:knqrziiarnfvthssuwksokqrnm

A penalty function method for exploratory adaptive-critic neural network control

Gianluca Di Muro, Silvia Ferrari
2009 2009 17th Mediterranean Conference on Control and Automation  
A constrained penalty function method for exploratory adaptive-critic neural network (NN) control is presented.  ...  While constrained approximate dynamic programming has been effective to guarantee closed-loop system performance and stability objectives, in the presence of a change in the plant dynamics it may not have  ...  Forgetting: exploratory adaptive function approximation Suppose that an Artificial Neural Network (ANN) is composed of Short Term Memory (STM) and Long Term Memory (LTM) connections and that it has been  ... 
doi:10.1109/med.2009.5164744 fatcat:z76gzjek5vf2zcoxq6x7quy73a

Nonlinear adaptive flight control using neural networks

1998 IEEE Control Systems  
Feedback linearization and adaptive neural networks provide a powerful controller architecture.  ...  A description of the controller architecture and associated stability analysis is given, followed by a more in-depth look at its application to a tiltrotor aircraft.  ...  This research supported by ARO, AFWL, AFOSR, and NASA. The authors gratefully acknowledge the feedback received from anonymous reviewers.  ... 
doi:10.1109/37.736008 fatcat:5horwo6ry5fmdfg2qa7ngrtcvm

Fuzzy PD Control of Networked Control Systems Based on CMAC Neural Network

Li-lian Huang, Jin Chen
2012 Mathematical Problems in Engineering  
The cerebellar model articulation controller (CMAC) neural network and a PD controller are combined to achieve the forward feedback control.  ...  The network and plant can be regarded as a controlled time-varying system because of the random induced delay in the networked control systems.  ...  Acknowledgments This work is supported by the Chinese National Science Foundation no. 61203004 , the Natural Science Foundation of Heilongjiang province no. 42400621-1-12201 , and the Fundamental Research  ... 
doi:10.1155/2012/347217 fatcat:srnrfoog3nhhtixmtbyrj7xiua

Flight Testing of Reconfigurable Control Law on the X-36 Tailless Aircraft

Joseph S. Brinker, Kevin A. Wise
2001 Journal of Guidance Control and Dynamics  
The reconfigurable control law employs an adaptive neural network to augment a dynamic inversion control law design.  ...  Real-time performance and long-term stability of the adaptive neural network based reconfigurable con- trol law was demonstrated through the HILS and flight testing ef- forts.  ... 
doi:10.2514/2.4826 fatcat:wpswomi6gngejlsgiut3w3kkxm

Implementation of an Adaptive Robust Neural Network Based Motion Controller for Position Tracking of AC Servo Drives

Won-Ho Kim
2009 International Journal of Fuzzy Logic and Intelligent Systems  
The neural network with radial basis function is introduced for position tracking control of AC servo drive with the existence of system uncertainties.  ...  The stability and the convergence of the system are proved by Lyapunov theory. The validity and robustness of the controller are verified through simulation and experimental results  ...  Adaptive Robust Neural Network Controller This section introduces the design of proposed controller for the AC servo drive and the proof of stability.  ... 
doi:10.5391/ijfis.2009.9.4.294 fatcat:anktkiw7rnauhhrxtt2xbmxdgy

Missile Control Using Fuzzy Cerebellar Model Arithmetic Computer Neural Networks

Z. Jason Geng, Claire L. McCullough
1997 Journal of Guidance Control and Dynamics  
More importantly, the missile dy- namics is stabilized before the fuzzy CMAC learning process be- gins, which makes the neural network easier to adapt and results in high performance.  ...  The inner feedback control level guarantees the stability of the flight control system, and the fuzzy CMAC adaptive learning controller enhances the performance and reliability by compensating for the  ... 
doi:10.2514/2.4077 fatcat:hltjtnedyfbwnkzlxrhbpfgave

Nonlinear Flight Control Using Neural Networks

Byoung S. Kim, Anthony J. Calise
1997 Journal of Guidance Control and Dynamics  
Simulation results for an F-18 aircraft model are presented to illustrate the performance of the on-line neural network based adaptation algorithm.  ...  The theoretical development of a direct adaptivetracking control architecture using neural networks is presented.  ...  Acknowledgments This research was supported by the Naval Air Warfare Center under Contract N62269-91-C-0038,and by the Army Research Ofce under Grant DAAH04-93-G-0002. The authors are grateful to P.  ... 
doi:10.2514/2.4029 fatcat:kdkhr4smgjci3ebjpjltcupyhm

An adaptive critic global controller

S. Ferrari, R.F. Stengel
2002 Proceedings of the 2002 American Control Conference (IEEE Cat. No.CH37301)  
A nonlinear control system comprising a network of networks is taught using a two-phase learning procedure realized through novel techniques for initialization, on-line training, and adaptive critic design  ...  The result is an adaptive controller that is as conservative as the linear designs and as effective as the global controller.  ...  The neural control architecture specified in Section 4 is initialized based on the performance criteria established locally by the linear gains and the augmented Riccati matrix, with NNF approximating  ... 
doi:10.1109/acc.2002.1025189 fatcat:avl5bmtpnbhv3fswfu64nwuy4e

Filtering and Control for Unreliable Communication: The Discrete-Time Case

Guoliang Wei, Lifeng Ma, Zidong Wang
2014 Discrete Dynamics in Nature and Society  
The global exponential stability analysis of discrete-time neural networks is discussed in "Global exponential stability of discrete-time neural networks with time-varying delays" by S. Udpin and P.  ...  Niamsup, and some global stability criteria of discrete-time neural networks with time-varying delays are proposed.  ...  Acknowledgments This special issue is a timely reflection of the recent research progress in the area of filtering and control for unreliable communication in the discrete-time domain.  ... 
doi:10.1155/2014/890275 fatcat:6dettkhmxrhhhdreo6flwkr4pa

Decentralized model reference adaptive control without restriction on subsystem relative degrees

Changyun Wen, Yeng Chai Soh
1999 IEEE Transactions on Automatic Control  
When the direct model reference adaptive control (MRAC) scheme with first-order local estimators is employed to design totally decentralized controllers, the stability result can only be applied to a system  ...  With this analysis, the class of interactions and subsystem unmodeled dynamics can be enlarged to include those having infinite memory.  ...  STABILITY OF THE DECENTRALIZED ADAPTIVE CONTROL SYSTEMS We need to establish the robustness of the local adaptive controllers in the presence of ignored interactions, unmodeled dynamics, and external disturbances  ... 
doi:10.1109/9.774124 fatcat:f3upzpxmxbb75bb2si6hibrkty
« Previous Showing results 1 — 15 out of 16,472 results