Robust adaptive backstepping fast terminal sliding mode controller for uncertain quadrotor UAV

Prediction based on time series has a wide range of applications. Due to the complex nonlinear and random distribution of time series data, the performance of learning prediction models can be reduced by the modeling bias or overfitting. This paper proposes a novel planar flow-based variational auto-encoder prediction model (PFVAE), which uses the long- and short-term memory network (LSTM) as the auto-encoder and designs the variational auto-encoder (VAE) as a time series data predictor to overcome the noise effects. In addition, the internal structure of VAE is transformed using planar flow, which enables it to learn and fit the nonlinearity of time series data and improve the dynamic adaptability of the network. The prediction experiments verify that the proposed model is superior to other models regarding prediction accuracy and proves it is effective for predicting time series data.

/pdf/pfvae-a-planar-flow-based-variational-auto-encoder-1igchrd1.pdf

PFVAE: A Planar Flow-Based Variational Auto-Encoder Prediction Model for Time Series Data

The purpose of this paper is to solve the problem of controlling of the quadrotor exposed to external constant disturbances. The quadrotor system is partitioned into two parts: the attitude subsystem and the position subsystem. A new robust integral terminal sliding mode control law (RITSMC) is designed for stabilizing the inner loop and the quick tracking of the right desired values of the Euler angles. To estimate the disturbance displayed on the - axis and to control the altitude position subsystem, an adaptive backstepping technique is proposed, while the horizontal position subsystem is controlled using the backstepping approach. The stability of the quadrotor subsystems is guaranteed by the Lyapunov theory. The effectiveness of the proposed methods is clearly comprehended through the obtained results of the various simulations effectuated on MATLAB/Simulink, and a comparison with another technique is presented.

/pdf/robust-integral-terminal-sliding-mode-control-for-quadrotor-x3ruhb0ljt.pdf

Robust Integral Terminal Sliding Mode Control for Quadrotor UAV with External Disturbances

Quadrotors are highly maneuverable drones, which are susceptible to the parameter uncertainties such as the mass, drag coefficients, and moment of inertia. Whose nonlinearities, aerodynamic disturbances, and higher coupling between the rotational and the translational dynamics stand for a problem that demands a robust controller. In the present paper, a fractional order (FO) improved super twisting proportional-integral-derivative sliding-mode control (STPIDSMC) is proposed for the quadrotor system. To improve the speed tracking performance, a FOPIDSM surface is designed. Moreover, the proposed FO control approach ensures fast convergence, high precision, good robustness against stochastic perturbations and uncertainties. Finally, the performance of the FOSTPIDSMC is investigated under different scenarios. The simulation results clearly show the high control performance, efficiency and high disturbance rejection capacity of the controller strategy proposed in this work in comparison with the nonlinear internal model control (NLIMC) and FO backstepping sliding mode control (FOBSMC) strategies.

Path Following Control of Quadrotor UAV With Continuous Fractional-Order Super Twisting Sliding Mode

Deep learning (DL)-based PV Power Forecasting (PVPF) emerged nowadays as a promising research direction to intelligentize energy systems. With the massive smart meter integration, DL takes advantage of the large-scale and multi-source data representations to achieve a spectacular performance and high PV forecastability potential compared to classical models. This review article taxonomically dives into the nitty-gritty of the mainstream DL-based PVPF methods while showcasing their strengths and weaknesses. Firstly, we draw connections between PVPF and DL approaches and show how this relation might cross-fertilize or extend both directions. Then, fruitful discussions are conducted based on three classes: discriminative learning, generative learning, and deep reinforcement learning. In addition, this review analyzes recent automatic architecture optimization algorithms for DL-based PVPF. Next, the notable DL technologies are thoroughly described. These technologies include federated learning, deep transfer learning, incremental learning, and big data DL. After that, DL methods are taxonomized into deterministic and probabilistic PVPF. Finally, this review concludes with some research gaps and hints about future challenges and research directions in driving the further success of DL techniques to PVPF applications. By compiling this study, we expect to help aspiring stakeholders widen their knowledge of the staggering potential of DL for PVPF.

/pdf/convergence-of-photovoltaic-power-forecasting-and-deep-5ed6hrwvjm.pdf

Convergence of Photovoltaic Power Forecasting and Deep Learning: State-of-Art Review

Short-term self consumption PV plant power production forecasts based on hybrid CNN-LSTM, ConvLSTM models

CNN-LSTM: An efficient hybrid deep learning architecture for predicting short-term photovoltaic power production

Commonly, three Hall effect sensors are used to detect the rotor position in brushless motors. Since one of these sensors may fail and lead to unstable operation, fault-tolerant control (FTC) of Hall sensor drives has attracted renewed research attention recently to ensure very high reliability. This article presents simple combinatory functions to detect fails in Hall effect sensors based only on Hall sensors signals, which require just a memory of 3 b. The proposed method guarantees more reliability and overcomes erroneous alarms that can be triggered by previous fault diagnosis methods using the information on the observed acceleration, which may be the subject of unexpected variations and lead consequently to false diagnosis. In addition, the article opens the discussion about the BLDC startup with a faulty sensor, in order to perform FTC algorithms ensuring a long use of the motor given the difficulty to repair Hall effect sensors inserted between the coils inside the stator. To confirm the theoretical analysis, experimental results are reported and limitations on the performances of the faulty system in terms of fault detection and signal reconstruction algorithm are discussed. The proposed schemes circuits can be inserted in conventional drive systems or by being placed between the existing driver and the motor.

Binary Diagnosis of Hall Effect Sensors in Brushless DC Motor Drives

In the last years, many research studies have been presented on quadrotors due to their different applications in daily life. In this paper, a complete mathematical model of generalized flight is developed based on the Newton-Euler method. This one takes into account external disturbances (aero-dynamic forces and moments) acting on the vehicle. A nonlinear control strategy based on the sliding mode control technique has been designed. This controller is employed to control the drone attitude and position as well as to compensate for external disturbances acting on the quadrotor. The stability of the system in closed-loop is ensured by Lyapunov’s theorem and analysis. A control method integral sliding mode controller (ISMC) has been developed to improve the controller performance of the SMC. The problem of chattering in the suggested controllers was treated without degradation of the system performance (precision, rapidity, robustness). Finally, to validate the choice of controllers proposed, several simulations were developed with the presence of aerodynamic disturbances.

Modeling and Robust Integral Sliding Mode Control for a Quadrotor Unmanned Aerial Vehicle

Isn’t it wonderful to see a floating body in the sky? Challenging multiple physics laws with remarkable control and stability. These are today’s flying vehicles. In this paper, the secret of this technique will be revealed differently. In the first place, the modeling of the system takes place using the Newton-Euler method allowing the balance of forces and moments on different parts of the manipulator, taking into account the disturbing factors acting on the aerial vehicle. It is therefore essential to design a controller (position and attitude) to ensure the stability of the floating body when following the desired set-point in the form of a 3D trajectory, a robust nonlinear controller based on the Lyapunov theory is designed. This latter has the ability to compensate the uncertain disturbing factors acting on the quadrotor. A comparative study of proportional-derivative (PD) and backstepping controllers comes to take place by comparing their performances using Matlab/Simulink as workspace environment to implement the plant model and different types of controllers.

Yassine El Houm

Papers

Short-term self consumption PV plant power production forecasts based on hybrid CNN-LSTM, ConvLSTM models

CNN-LSTM: An efficient hybrid deep learning architecture for predicting short-term photovoltaic power production

Binary Diagnosis of Hall Effect Sensors in Brushless DC Motor Drives

Modeling and Robust Integral Sliding Mode Control for a Quadrotor Unmanned Aerial Vehicle

A Comparative Analysis of Control Strategies for Stabilizing a Quadrotor