An adaptive neural control scheme is proposed for a class of generic hypersonic flight vehicles. ... more An adaptive neural control scheme is proposed for a class of generic hypersonic flight vehicles. The main advantages of the proposed scheme include the following: (1) a new constraint variable is defined to generate the virtual control that forces the tracking error to fall within prescribed boundaries; (2) RBF NNs are employed to compensate for complex and uncertain terms to solve the problem of controller complexity; (3) only one parameter needs to be updated online at each design step, which significantly reduces the computational burden. It is proved that all signals of the closed-loop system are uniformly ultimately bounded. Simulation results are presented to illustrate the effectiveness of the proposed scheme.
020 1st International Conference on Communications, Control Systems and Signal Processing (CCSSP)
The purpose of this paper, is to design a controller for quadrotor attitude system. The designed ... more The purpose of this paper, is to design a controller for quadrotor attitude system. The designed control law combines a continuous second-order sliding mode control (CSOSMC), the Fuzzy-Chebyshev network (FCN) and the adaptive control methodology. The FCN with adaptive parameters is exploited to approximate the nonlinear functions and improve the robustness against parametric uncertainties and external disturbances. Otherwise, the continuous sliding mode aims to completely eliminate the chattering phenomenon. The stability of the quadrotor attitude control system is proven by the Lyapunov stability approach. The simulation results demonstrate the capability and efficiency of the proposed technique in the presence of uncertainties and external disturbances.
020 1st International Conference on Communications, Control Systems and Signal Processing (CCSSP)
This work consists in designing a backstepping controller based on an adaptive fuzzy neural netwo... more This work consists in designing a backstepping controller based on an adaptive fuzzy neural network (FNN). The main aim is the attitude control of a quadrotor system under uncertainties and disturbances. The FNN with adaptive parameters is exploited to approximate the nonlinear functions and improve the robustness against parametric uncertainties and external disturbances. FNN is included in classical backstepping control (BC) to solve the unknown dynamics problem. Otherwise, a robust control term is added to improve performance in tracking a reference signal when parametric uncertainties and disturbances exist. The stability of the quadrotor attitude control system is proven by the Lyapunov method. Simulation results of the proposed adaptive fuzzy neural network based decentralized backstepping controller (AFNN-DBC) demonstrate the capability and efficiency of the proposed technique in the presence of uncertainties and external disturbances in comparison with classical backstepping controller (BC).
Advances in Differential Equations and Control Processes
An intelligent adaptive fractional-order backstepping control under unknown external disturbances... more An intelligent adaptive fractional-order backstepping control under unknown external disturbances and parameter uncertainties for Lemya Guettal, Riadh Ajgou and Mostefa Mohamed Touba 100 quadrotor is developed. The developed approach named FCN-FOBC combines fractional-order backstepping control (FOBC) and fuzzy-Chebyshev network (FCN). Initially, the overall control and system tracking are performed using backstepping control (BC). FOBC is designed to advance the convergence speed and control reliability. Second, the FCN is set up to approximate the uncertainties, and a robust term is considered to overcome the problem of FCN approximation errors. Finally, using the Lyapunov theory, the stability of control system is confirmed. The numerical results confirm that the proposed controller has better tracking accuracy and stronger robustness compared to conventional approaches.
020 1st International Conference on Communications, Control Systems and Signal Processing (CCSSP), 2020
Flying within highly unstructured and unknown environments performs a big confrontation for gener... more Flying within highly unstructured and unknown environments performs a big confrontation for general understanding of the view on-board a future platform for unmanned aerial vehicles (UAV) technology. In this work, an approach for automatic trail management within forest trail environment is presented. Our proposed approach generalizes successfully with high image resolutions allowing the UAV to manage in difficult natural conditions. A deep neural network is optimized to improve detection of forest paths for guiding UAVs, by using images of sufficiently high-resolution that are representative of UAV platform.
In this paper, robust feature extraction and efficient speech activity detection algorithm are pr... more In this paper, robust feature extraction and efficient speech activity detection algorithm are proposed for improvement in remote speaker recognition system over AWGN (Additive White Gaussian Noise) channel. Moreover, The system employs a robust speech feature based on AR-MFCC modeled with GMM model and applying an efficient speech activity detection (SAD) algorithm with adaptive threshold. Furthermore, the proposed speech activity detection algorithm is based on Zero Crossing Rate and Energy Measurements with prior SNR estimation by estimating noise. variance where the algorithms were implemented and tested in MATLAB. Besides, Feature extraction requires much attention because recognition performance depends heavily on this phase. The Mel-Frequency Cepstral coefficient (MFCC) is a very useful feature for speaker recognition in clean conditions but it deteriorates in the presence of noise. Thus, in our work, feature extraction framework based on the combination of MFCC and Autoregressive model (AR) parameters has been proposed. , the TIMIT database with speech from 630 speakers has been used in MATLAB simulation. The first four utterances for each speaker could be defined as the training set while 1 utterance as the test set. The use of AR-MFCC approach has provided significant improvements in identification rate accuracy when compared with MFCC. However, in terms of runtime, AR-MFCC requires more time to execute than MFCC
An adaptive neural control scheme is proposed for a class of generic hypersonic flight vehicles. ... more An adaptive neural control scheme is proposed for a class of generic hypersonic flight vehicles. The main advantages of the proposed scheme include the following: (1) a new constraint variable is defined to generate the virtual control that forces the tracking error to fall within prescribed boundaries; (2) RBF NNs are employed to compensate for complex and uncertain terms to solve the problem of controller complexity; (3) only one parameter needs to be updated online at each design step, which significantly reduces the computational burden. It is proved that all signals of the closed-loop system are uniformly ultimately bounded. Simulation results are presented to illustrate the effectiveness of the proposed scheme.
020 1st International Conference on Communications, Control Systems and Signal Processing (CCSSP)
The purpose of this paper, is to design a controller for quadrotor attitude system. The designed ... more The purpose of this paper, is to design a controller for quadrotor attitude system. The designed control law combines a continuous second-order sliding mode control (CSOSMC), the Fuzzy-Chebyshev network (FCN) and the adaptive control methodology. The FCN with adaptive parameters is exploited to approximate the nonlinear functions and improve the robustness against parametric uncertainties and external disturbances. Otherwise, the continuous sliding mode aims to completely eliminate the chattering phenomenon. The stability of the quadrotor attitude control system is proven by the Lyapunov stability approach. The simulation results demonstrate the capability and efficiency of the proposed technique in the presence of uncertainties and external disturbances.
020 1st International Conference on Communications, Control Systems and Signal Processing (CCSSP)
This work consists in designing a backstepping controller based on an adaptive fuzzy neural netwo... more This work consists in designing a backstepping controller based on an adaptive fuzzy neural network (FNN). The main aim is the attitude control of a quadrotor system under uncertainties and disturbances. The FNN with adaptive parameters is exploited to approximate the nonlinear functions and improve the robustness against parametric uncertainties and external disturbances. FNN is included in classical backstepping control (BC) to solve the unknown dynamics problem. Otherwise, a robust control term is added to improve performance in tracking a reference signal when parametric uncertainties and disturbances exist. The stability of the quadrotor attitude control system is proven by the Lyapunov method. Simulation results of the proposed adaptive fuzzy neural network based decentralized backstepping controller (AFNN-DBC) demonstrate the capability and efficiency of the proposed technique in the presence of uncertainties and external disturbances in comparison with classical backstepping controller (BC).
Advances in Differential Equations and Control Processes
An intelligent adaptive fractional-order backstepping control under unknown external disturbances... more An intelligent adaptive fractional-order backstepping control under unknown external disturbances and parameter uncertainties for Lemya Guettal, Riadh Ajgou and Mostefa Mohamed Touba 100 quadrotor is developed. The developed approach named FCN-FOBC combines fractional-order backstepping control (FOBC) and fuzzy-Chebyshev network (FCN). Initially, the overall control and system tracking are performed using backstepping control (BC). FOBC is designed to advance the convergence speed and control reliability. Second, the FCN is set up to approximate the uncertainties, and a robust term is considered to overcome the problem of FCN approximation errors. Finally, using the Lyapunov theory, the stability of control system is confirmed. The numerical results confirm that the proposed controller has better tracking accuracy and stronger robustness compared to conventional approaches.
020 1st International Conference on Communications, Control Systems and Signal Processing (CCSSP), 2020
Flying within highly unstructured and unknown environments performs a big confrontation for gener... more Flying within highly unstructured and unknown environments performs a big confrontation for general understanding of the view on-board a future platform for unmanned aerial vehicles (UAV) technology. In this work, an approach for automatic trail management within forest trail environment is presented. Our proposed approach generalizes successfully with high image resolutions allowing the UAV to manage in difficult natural conditions. A deep neural network is optimized to improve detection of forest paths for guiding UAVs, by using images of sufficiently high-resolution that are representative of UAV platform.
In this paper, robust feature extraction and efficient speech activity detection algorithm are pr... more In this paper, robust feature extraction and efficient speech activity detection algorithm are proposed for improvement in remote speaker recognition system over AWGN (Additive White Gaussian Noise) channel. Moreover, The system employs a robust speech feature based on AR-MFCC modeled with GMM model and applying an efficient speech activity detection (SAD) algorithm with adaptive threshold. Furthermore, the proposed speech activity detection algorithm is based on Zero Crossing Rate and Energy Measurements with prior SNR estimation by estimating noise. variance where the algorithms were implemented and tested in MATLAB. Besides, Feature extraction requires much attention because recognition performance depends heavily on this phase. The Mel-Frequency Cepstral coefficient (MFCC) is a very useful feature for speaker recognition in clean conditions but it deteriorates in the presence of noise. Thus, in our work, feature extraction framework based on the combination of MFCC and Autoregressive model (AR) parameters has been proposed. , the TIMIT database with speech from 630 speakers has been used in MATLAB simulation. The first four utterances for each speaker could be defined as the training set while 1 utterance as the test set. The use of AR-MFCC approach has provided significant improvements in identification rate accuracy when compared with MFCC. However, in terms of runtime, AR-MFCC requires more time to execute than MFCC
Uploads
Papers by Lemya Guettal