An SRWNN-based approach on developing a self-learning and self-evolving adaptive control system for motion platforms

Ari, Evrim Onur
Kocaoglan, Erol
In this paper, a self-recurrent wavelet neural network (SRWNN)-based indirect adaptive control architecture is modified for performing speed control of a motion platform. The transient behaviour of the original learning algorithm has been improved by modifying the learning rate updates. The contribution of the proposed modification has been verified via both simulations and experiments. Moreover, the performance of the proposed architecture is compared with robust RST designs performed on a similar benchmark system, to show that via adaptive nonlinear control, it is possible to obtain a fast step response without degrading the robustness of a multi-body mechanical system. Finally, the architecture is further improved so as to possess structural learning for populating the SRWNNs automatically, rather than employing static network structures, and simulation results are provided to show the performance of the proposed structural learning algorithm.


An experimental study on Power Amplifier linearisation by artificial neural networks Yapay Sinir Aǧlari ile Güç Yükselteç Doǧrusalląstirma Amaçli Deneysel Bir Çalisma
Yesil, Soner; Kolagasioglu, Ahmet Ertugrul; Yılmaz, Ali Özgür (2018-07-05)
This paper represents an experimental study on the linearisation of Power Amplifiers especially on high output power regions by utilizing an artificial neural network structure and open-loop training method. For the same in-band output power, 9dB EVM and 6dB ACLR improvement has been observed on hardware by feeding the proposed digital predistortion signal (DPD) to the PA under test.
A Control System Architecture for Control of Non-Affine in Control, Open-Loop Unstable Underactuated Systems
Marangoz, Alp; Kutay, Ali Türker (2017-07-25)
In this paper, a control system architecture for control of non-affine in control, open-loop unstable underactuated system is discussed. Passivization of the unactuated (internal) system dynamics achieved through perturbation of trajectories of the actuated states, which are calculated through adaptive dynamic inversion technique, based on Tikhonov's theorem. Performance of the controller is shown through simulation of two open-loop unstable and locally uncontrollable example problems.
An adaptive, energy-aware and distributed fault-tolerant topology-control algorithm for heterogeneous wireless sensor networks
Deniz, Fatih; Bagci, Hakki; KÖRPEOĞLU, İBRAHİM; Yazıcı, Adnan (2016-07-01)
This paper introduces an adaptive, energy-aware and distributed fault-tolerant topology control algorithm, namely the Adaptive Disjoint Path Vector (ADPV) algorithm, for heterogeneous wireless sensor networks. In this heterogeneous model, we have resource-rich supernodes as well as ordinary sensor nodes that are supposed to be connected to the supernodes. Unlike the static alternative Disjoint Path Vector (DPV) algorithm, the focus of ADPV is to secure supernode connectivity in the presence of node failures...
A General framework on adaptive hybrid beamformingand channel acquisition for wideband mm-wave massive MIMO systems
Kurt, Anıl; Güvensen, Gökhan Muzzaffer.; Department of Electrical and Electronics Engineering (2019)
In this thesis, an efficient hybrid beamforming architecture together with a novel spatio-temporal receiver processing is proposed for single-carrier (SC) mm-wave wideband massive MIMO channels in time-domain duplex (TDD) mode. The design of two-stage beamformer is realized by using a virtual sectorization via second-order channel statistics based user grouping. The novel feature of the proposed architecture is that the effect of both inter-group-interference (due to non-orthogonality of virtual angular sec...
Representing temporal knowledge in connectionist expert systems
Alpaslan, Ferda Nur (1996-09-27)
This paper introduces a new temporal neural networks model which can be used in connectionist expert systems. Also, a Variation of backpropagation algorithm, called the temporal feedforward backpropagation algorithm is introduced as a method for training the neural network. The algorithm was tested using training examples extracted from a medical expert system. A series of experiments were carried out using the temporal model and the temporal backpropagation algorithm. The experiments indicated that the alg...
Citation Formats
E. O. Ari and E. Kocaoglan, “An SRWNN-based approach on developing a self-learning and self-evolving adaptive control system for motion platforms,” INTERNATIONAL JOURNAL OF CONTROL, pp. 380–396, 2016, Accessed: 00, 2020. [Online]. Available: