Search

The attitude synchronization problem for multiple spacecraft with angular velocity and input constraints is investigated in this article, under the mild assumptions that the leader's states can reach each follower through a path and the communications between followers are bidirectional. An adaptive finite-time distributed observer is designed to estimate the leader's states for each follower spacecraft, which has the superiority in the fully distributed that the communication topology or some other global information is not required, and the finite-time convergence can be guaranteed. Furthermore, the novel auxiliary systems are first proposed to avoid the angular velocity and input constraints being violated, while the finite-time command filtered backstepping controller is designed to track the leader's attitude motion. The remarkable features of the developed algorithm are the global finite-time attitude consensus of multiple spacecraft, with the constraints satisfied. Moreover, the efficiency of the proposed method is illustrated by numerical simulations and real-time platform verification.

In this paper, a novel composite controller design technique is addressed for using an adaptive interval type-2 fuzzy logic system (FLS) with the fixed-time disturbance observer (FTDO) such that the enhanced tracking performance is achieved for an air-breathing hypersonic vehicle with a variable geometry inlet (AHV-VGI). First, introducing the variable geometry inlet to take proper mass flow into the engine without any flow spillage ensures enough thrust for acceleration and maneuvering flight. The interval type-2 FLS integrated with an adaptive control algorithm is derived to approximate the complicated nonlinear items in the control strategy online, which assures that the tracking error are semi-globally uniformly bounded. After that, the fixed-time disturbance observer that is constructed to provide the estimations of uncertainty including external disturbance leads to a more practical and robust flight control system for AHV-VGI. The uniform stability of the whole system is proved under the framework of Lyapunov theory. Finally, simulation results and analysis are presented to demonstrate the effectiveness of the proposed composite control scheme.

As one of the tiniest flying objects, unmanned aerial vehicles (UAVs) are often expanded as the ``swarm'' to execute missions. In this article, we investigate the multiquadcopter and target pursuit-evasion game in the obstacles environment. For high-quality simulation of the urban environment, we propose the pursuit-evasion scenario (PES) framework to create the environment with a physics engine, which enables quadcopter agents to take actions and interact with the environment. On this basis, we construct multiagent coronal bidirectionally coordinated with target prediction network (CBC-TP Net) with a vectorized extension of multiagent deep deterministic policy gradient (MADDPG) formulation to ensure the effectiveness of the damaged ``swarm'' system in pursuit-evasion mission. Unlike traditional reinforcement learning, we design a target prediction network (TP Net) innovatively in the common framework to imitate the way of human thinking: situation prediction is always before decision-making. The experiments of the pursuit-evasion game are conducted to verify the state-of-the-art performance of the proposed strategy, both in the normal and antidamaged situations.

The finite-time attitude maneuvering control and vibration suppression for the flexible spacecraft with external disturbances, inertia uncertainties and input saturation are investigated in this paper. Firstly, the input shaping technique is applied to suppress most of the vibration caused by flexible appendages. Then, for the nominal attitude control system, a multivariable continuous sliding mode controller is applied to guarantee the finite-time convergence. Next, unlike the conventional discontinuous design of integral sliding mode, a continuous compensated controller with modified multivariable adaptive twisting algorithm is proposed to reject the lumped uncertainty, including disturbances, uncertainties, saturation and residual vibration. The controller can realize finite time convergence and the chattering suppression, while there’s no need for the priori knowledge of the lumped uncertainty. A rigorous proof of the finite time stability of the closed-loop system is derived by the Lyapunov method. Finally, the efficiency of the proposed method is illustrated by numerical simulations.

The tracking control of a flexible hypersonic vehicle with system perturbation and actuator fault is investigated in this paper. A control-oriented model is constructed, which is composed of velocity loop and altitude loop. A double-layer fast adaptive-gain super-twisting disturbance observer (ASTDO) is designed to provide finite-time estimation of the lumped terms of external disturbances and actuator faults. The ASTDO has the property of the non-overestimation of adaptive gains and chattering reduction with higher precision. Based on the estimation of ASTDO, a novel finite-time command filtered backstepping (FCFB) scheme is proposed for the high-order altitude loop. The merit lies in the modified compensating signals and new virtual commands to compensate the command filter errors. The remarkable features of the developed algorithm include the finite-time convergence with higher accuracy and the chattering suppression. Finally, comparative simulation results are given to demonstrate the superiority of the proposed method.

This paper investigates the problem of attitude tracking control for reusable launch vehicles (RLVs) with aeroservoelastic characteristic and disturbances in reentry phase. An aeroservoelastic model for reentry RLV is first established, where the synthetic disturbances are considered as well. Then a novel disturbance observer combined with fuzzy logic system is constructed by using both fuzzy approximation error and observation error to provide the disturbance estimation. The fuzzy disturbance observer (FDO) can be employed to the control scheme to counteract the perturbations and guarantee the estimation errors for converging to a small neighborhood around origin. Incorporated with the designed FDOs, the adaptive sliding mode technique is proposed for developing attitude angle and angular rate subsystem controllers, respectively, to ensure the reentry attitude tracking performance. Rigorous proof shows that the uniform stability of the closed-loop system under the proposed control law can be guaranteed by using Lyapunov technique. Finally, numerical simulations are conducted in this paper to illustrate the effectiveness of the developed technique for the RLV.

In this paper, we aim to tackle the formation tracking control for a Quadrotor UAV system using the adaptive dynamic programming. The Quadrotor UAV system is composed by position outer-loop and attitude inner-loop. The main contribution is that an adaptive dynamic programming based robust tracking control scheme is established for the outer-loop position subsystem with external disturbances and model uncertainty. Then the tracking signal of the attitude subsystem is obtained by attitude resolution, and a sliding mode controller is designed to get fast tracking. The robust tracking control performance is guaranteed theoretically via Lyapunov approach and also verified through simulation illustration.

Unmanned aerial vehicle (UAV) swarms have attracted great interests in numerous civil and military scenarios due to their flexible deployment and mobility. The increased network size poses a challenge for the network scalability. Clustering as an effective scheme, is able to provide efficient communication and network performance. In order to adapt to the change of network topology, the link subsistence probability is used to evaluate the mobility of UAVs. However, the UAVs usually perform tasks collaboratively, which leads to their movements are typically group-based. To this end, a coalition game theoretic framework is proposed to cluster UAVs into coalitions in a distributed autonomous manner, based on the group and the mobility information. The proposed game allows the UAVs with the same group, as much as possible, into one coalition under some constraints, such as cluster size and cluster diameter. To solve the game, each UAV makes its decision whether to switch based on the coalition value. Finally, some comparisons are provided to illustrate the efficiency of the proposed algorithm.

This paper investigates a fuzzy disturbance observer (FDO)-based terminal sliding mode control (TSMC) strategy for the liquid-filled spacecraft with flexible structure(LFS-FS) under control saturation. Firstly, a novel FDO is designed to estimate the lumped uncertainty, including the inertia uncertainty, external disturbance, the coupling of liquid slosh and flexible structure(LF), as well as the parts that exceed control saturation. The merits of the FDO lie in that estimation error can be arbitrarily small by adjusting the designed parameters and the prior information is not required. Then, based on the estimation of FDO, a finite-time TSMC is designed, which has more satisfactory control performance, such as chattering reduction and fast convergence speed. The stability of the closed-loop system is proved strictly by Lyapunov theory. Finally, numerical simulations are presented to demonstrate the effectiveness of the proposed method.

A safe and rapid formation generation is of great importance for the cooperation performance of the multiple unmanned helicopters. The trajectory optimization and control is the key problem in the process of the formation generation. To achieve this goal, a novel safe formation generation strategy is proposed through trajectory optimization and tracking control. First, the rapidly-exploring random tree (RRT) algorithm is applied to obtain the initial guess to satisfy the requirement of rapid solution. Then, the Gauss pseudospectral method is used to transform the optimal control problem to the nonlinear programming (NLP) problem and the sequence quadratic programming (SQP) method is utilized to gain the optimal trajectory combined with the initial guess. Second, a finite-time sliding mode controller is proposed to ensure the finite time trajectory tracking in the presence of model parameter uncertainties and unknown external disturbances. Finally, the numerical simulation is provided to show the effectiveness of the proposed formation generation strategy for unmanned helicopters formation.

In recent years, an increasingly popularity of deep learning model for intelligent state monitoring, diagnosis and prediction of spacecraft has been observed. However, in the previous studies, a major assumption accepted by default is that source domain data and target domain data have same feature distribution. Unfortunately, this assumption is mostly invalid in real application. Considering the problem that the original fault data sample is small, the noise is high and the fault signal is unlabeled, in this paper, we propose deep transfer learning-based fault diagnosis method for spacecraft system in which a new fault diagnosis framework-deep transfer network(DTN) is built, and it can generalize deep learning model to domain adaptation scenario inspired by the idea of transfer learning. In order to improve the accuracy of on-orbit spacecraft fault data detection, the proposed framework with joint distribution adaptation(JDA) is applied to exploit the distribution structure of unlabeled data in target domain by using the data with labels in source domain. By comparing with other methods, the deep transfer network based on joint distribution adaptation has better transfer performance in fault diagnosis of spacecraft.

Multiple unmanned aerial vehicles (UAVs) are applicable to numerous civil and military scenarios, such as rescue, surveillance, inspection, mapping, and so on. In fact, there exists lifetime and distance restrictions in UAVs due to the limited energy and signal attenuation. Therefore, a distributed algorithm is proposed which makes use of cooperative communication to ensure reliable information transmission of UAVs in communication-limited environments. First, a system model is developed by jointly considering the achievable rate, delay and energy consumption, which constitute the utility function of UAVs. Then, in order to study the interactions among UAVs, we present a game framework which allows the UAVs making decision under the global topology. Via a hybrid of the pure-strategy and the mixed-strategy Nash network formation algorithm, a multi-hop tree structure network that connects the UAVs and the ground station is established. Based on our numerical simulation, it demonstrates that the UAVs can adapt to the changing environment. Finally, some comparisons are provided to illustrate the efficiency of the proposed algorithm.

In this paper, a multi-model switching control is developed for air-breathing hypersonic vehicle with variable geometry inlet(AHV-VGI). A variable geometry inlet with the translating cowl is adopted to capture the enough air mass flow for the scramjet engine, which can ensure a more powerful thrust. However, the using of VGI causes the unknown changes of the aerodynamics and thrust, making the model of AHV more complex. Therefore, we firstly analyze the thrust characteristic with the translating cowl and present the conception of optimal elongation distance of translating cowl(EDTC). Consequently, multiple different nonlinear aerodynamic models are constructed by curve fitting for each position of the translating cowl. Then, a switching mechanism dependent on EDTC is proposed and the adaptive RBF neural controllers are designed for velocity subsystem and altitude system of every model. Furthermore, the common Lyapunov functional is constructed to prove the stability of the multi-model switching process. Finally, numerical simulations are given to demonstrate the effectiveness of the proposed control approach for AHV-VGI.

In this paper, the attitude control problem of reusable launch vehicles (RLVs) during reentry phase is investigated by using compound adaptive fuzzy H-infinity control (CAFHC) strategy in the presence of parameter uncertainties and external disturbances. Firstly, the control-oriented attitude model is established by a model transformation based on the six-degree-of-freedom (6-DoF) dynamic model of the RLV. Secondly, a novel attitude control scheme is developed and the control strategy consists of two parts to achieve a stable and accurate attitude tracking during reentry flight process. An attitude tracking controller is designed utilizing adaptive fuzzy H-infinity control approach combined with an identification model to improve the attitude tracking performance in the interior of fuzzy approximation region of attitude angle. Next, an attitude stabilization controller based on boundary adaptive technique is employed to assure the robustness of the closed-loop system in the exterior of fuzzy approximation region of attitude angle. Furthermore, the stability of the closed-loop system is guaranteed within the framework of Lyapunov theory and the attitude tracking error converges to a small neighborhood around origin. Finally, the simulation results are presented to demonstrate that the effectiveness of the proposed control scheme for reentry RLV, and its tracking performance performs better than the other control method.

For an air-breathing hypersonic vehicle with a variable geometry inlet (AHV-VGI), a movable translating cowl is usedto track the shock on lip conditions to capture enough air mass flow, which can extend the velocity range and be favorableto the acceleration and maneuvering flight. We firstly establish longitudinal dynamics for AHV-VGI, and consider the lumped disturbances which include the unknown external disturbance, the parameter uncertainties and the uncertainty parts introduced by translating cowl. Then, the dynamic surface control (DSC) strategy based on fuzzy disturbance observer (FDO) is proposed for AHV-VGI control. The control process for AHV-VGI is divided into two subsystems. For each subsystem, a sliding mode controller is designed, and FDOs are adopt to compensate the lumped disturbances, which can render the disturbance estimate errors convergent. Numerical simulations are presented to illustrate the effectiveness of the proposed method and the advantages of translating cowl.

This paper describes the design of attitude control for reusable launch vehicles during reentry phase using adaptive fuzzy control strategy with compensation controller that assures a stable and accurate attitude tracking despite having parameter uncertainties and external disturbances in the plant model. Firstly, the six-degrees-of-freedom dynamic model of the reusable launch vehicles is established, followed by a model transformation of rotational and kinematic dynamic equations, which results in a strict-feedback form attitude control system. Secondly, a fuzzy logic system combined with the adaptive technique is employed to model the system uncertainty term online. Then an attitude tracking controller based on adaptive fuzzy control approach is designed to assure tracking of guidance commands. Furthermore, to attenuate the adverse effects of fuzzy modeling errors on the control performance and system stability, a compensation controller is introduced to the control strategy. In addition, the stability of the closed-loop system is proved by using the Lyapunov theory, and the attitude tracking error converges to a small neighborhood of the origin. Finally, the simulation results and discussions are provided to verify the effectiveness of the proposed control strategy in tracking the guidance commands.

This paper presents a step-switching coordinate strategy for Air-breathing hypersonic vehicle with variable-geometry scramjet inlet(AHV-VGI), which can improve the aerodynamic performance of hypersonic vehicle by moving the movable lip to capture more airflow. Firstly, we transform the longitudinal model of AHV-VGI into the strict feedback form and multiple nonlinear aerodynamic models are built with different elongation distance of translating cowl(EDTC). Secondly, the longitudinal model of AHV-VGI system with different EDTC is divided into the velocity subsystem and altitude subsystem, which is designed by a finite-time nonsingular terminal sliding mode control method. Then, the common Lyapunov function is constructed to prove the stability of the closed-looped switching system with the control algorithms. Finally, simulation results are provided to validate the effectiveness of these theoretical results.

In this paper, an adaptive state observer for multiple quadrotors is designed to observe the relative state between the leader and followers. The upper bound of disturbance need not to be known in advance in the design of state observer. In the case of unknown disturbance upper bound, an adaptive disturbance observer for attitude loop is designed to estimate the complex disturbance of a quadrotor UAV, and the results of the disturbance observer are input into the controller to compensate the unknown disturbance in order to reduce the influence of these unfavorable factors on the stability of the system. In the design of attitude controller, it is difficult to obtain the derivative of desired attitude angle because of external disturbance and chattering caused by sliding mode control. Therefore, a third-order integral filter is designed between the two subsystems to estimate the desired attitude angle and its derivative, so as to avoid the influence of external interference and sliding mode chattering on the stability of the system.

This paper studied the problem of leader-following formation tracking control for multiple quadrotor unmanned aerial vehicles (UAV). The control object is to maintain the positions of follower quadrotors in formation with external disturbances and inner model parameter uncertainty. At the same time, only part of quadrotor UAV is informed to have access to the leader’s position. Firstly, a second order robust exact differentiator (2-RED) observer is proposed. By using continues high-order sliding mode technique, the output feedback controller is designed to position outer-loop based on the result of the outer-loop observer. Secondly, the output feedback controller and 2-RED observer are designed for attitude inner-loop. The finite-time stability of outer-loop and inner-loop controllers with observers are guaranteed through Lyapunov theories and homogeneity. Finally, numerical examples are presented to verify the developed algorithm.

This paper proposed a flexible vibration suppression strategy based on positive position feedback (PPF) for spacecraft with flexible structure. The PPF structure uses piezoelectric ceramic elements as sensors and actuators, and the vibration damping of the flexible mode is increased to achieve the purpose of active vibration suppression. From the perspective of system structure, the PPF compensator changes the original pole position of the system through the control algorithm, so as to increase the damping of the system. Combined with terminal sliding mode (TSM) adaptive controller, PPF strategy can actively suppress the vibrations of flexible accessories of spacecraft.

We propose a line/plane feature-based lidar inertial odometry and mapping method, for real-time 6-DOF pose estimation with unmanned ground vehicle. The accuracy of the estimated pose is mainly due to the correctness of the corresponding points matching. It is difficult with Velodyne-type sensors due to the sparse and non-uniform point clouds that they produce. In this paper, we describe a feature-based approach using segmentation and clustering algorithm, which results in mathematically principled line and plane features. A Levenberg-Marquardt optimization method then uses the line and plane features to solve different components of the six degree-of-freedom transformation across consecutive scans. We also use the pose-graph to perform loop-closure detection. We evaluate the performance of our algorithm by using datasets gathered from building environments with unmanned ground vehicles, and we compare our results against the state-of-the-art method LOAM. The results indicate that our method can achieve better accuracy.

A major challenge in motor imagery (MI) of electroencephalogram (EEG) based brain–computer interfaces (BCIs) is the individual differences for different people. That the classification model should be retrained from scratch for a new subject often leads to unnecessary time consumption. In this paper, a “brain-ID” framework based on the hybrid deep neural network with transfer learning (HDNN-TL) is proposed to deal with individual differences of 4-class MI task. An end-to-end HDNN is developed to learn the common features of MI signal. HDNN consists of convolutional neural network (CNN) and Long Short-Term Memory (LSTM) which are utilized to decode the spatial and temporal features of the MI signal simultaneously. To deal with the EEG individual differences problem, transfer learning technique is implemented to fine-tune the followed fully connected (FC) layer to accommodate new subject with fewer training data. The classification performance on BCI competition IV dataset 2a by the proposed HDNN-TL in terms of kappa value is 0.8. We compared HDNN-TL, HDNN and other state-of-art methods and the experimental results demonstrate that the proposed method can get a satisfying result for new subjects with less time and fewer training data in MI task.

This paper presents a novel fuzzy wavelet neural network structure to identify the nonlinear uncertain aeroelastic system. The identified aeroelastic system considers stiffness, damping non-linearity, dead zones and uncertainties. The proposed fuzzy wavelet neural network (FWNN) is developed from the interval type-2 fuzzy logic system which has the advantage to model uncertainties. Additionally, taking the rapidity and accuracy of the identification into account, the system is characterized by a set of fuzzy IF-THEN rules, and the consequent parts of which is designed to be single hidden layer wavelet neural network. And then, the sliding mode algorithm based on Lyapunov stability theory is introduced to obtain parameter update rules. Furthermore, numerical simulation for a structurally nonlinear prototypical two-dimensional wing section is investigated to verify the effectiveness of the proposed method. HighlightsThe identified AE system considers all of stiffness, damping non-linearity, dead zones and uncertainties.We propose a novel, fuzzy wavelet neural network (FWNN) structure.The structure is characterized by a set of fuzzy IF-THEN rules.The sliding mode algorithm based on Lyapunov stability theory is introduced to obtain parameter update rules.

This paper presents a fuzzy smooth-switching coordinate strategy for air-breathing hypersonic vehicle with variable geometry scramjet inlet (AHV-VGI), which can improve the aerodynamic performance of hypersonic vehicle by moving the movable lip to capture more airflow. The longitudinal model of AHV-VGI is firstly formulated, followed by a strict feedback form model. Wherein, the multiple nonlinear aerodynamic models are established with different elongation distance of translating cowl. Secondly, a finite-time nonsingular terminal sliding mode control (NTSMC) method is introduced to develop velocity and altitude subsystem controllers for each model. After that, an interval type-2 fuzzy logic system technique is employed to guarantee the smoothness during switching process between the adjacent NTSMC controllers. Furthermore, the Lyapunov function combined with adaptive laws are constructed to prove the finite-time stability of the closed-looped switching system. Finally, simulation results are provided to validate the effectiveness of these theoretical results.

This paper proposed a terminal sliding mode control strategy based on a type-2 fuzzy disturbance observer (FDO) for the spacecraft with liquid slosh and flexible structure (SLF). The type-2 fuzzy logic system (FLS) has a better behavior than type-l FLS on handling the uncertainties. Firstly, the type-2 FDO was adopted to estimate the whole uncertainties which contains the perturbations caused by the liquid fuel slosh and flexible structure, the unknown elements of the model and the external disturbances. Next, the terminal sliding mode surface was designed to ensure the system stability and robustness. Then, the stability of the proposed control scheme was proved based on the Lyapunov stability theory. Finally, numerical simulations showed the effectiveness and advantages of the proposed method.

This paper investigates finite-time formation tracking control problem for multiple quadrotors when the reference signal of the leader is not available to all the followers. In accordance with the multiple-timescale features, the dynamic model of quadrotor is divided into position subsystem and attitude subsystem. Firstly, distributed finite-time observers are designed to estimate the relative state errors between followers and the leader. Through Lyapunov theories, it is shown that the proposed observers can achieve finite-time stability. Secondly, based on the values of the observers, finite-time formation control algorithms including finite-time position controller and attitude controller are designed respectively. Thirdly, the finite-time stability analysis of the control algorithms is carried out using Lyapunov theory. Finally, simulation results are provided to validate the effectiveness of these theoretical results.

By using light-weighted material in hypersonic flight vehicle, the vehicle body can be easily deformed. The mutual couplings in aerodynamics, flexible structure and control system will result in aeroservoelasticity problem. Based on the six-DOF geometry of a hypersonic flight vehicle, this paper uses assumed modes method, Shock-Expansion theory and Piston theory to develop the structural elastic model and the aerodynamic model, with actuator incorporated, the open-loop aeroservoelastic model of a six-DOF HFV is established. In open-loop analysis, the response of the elastic coordinates to the tail rudder deflection and the frequency characteristic of the transfer function of the tail rudder deflection to the first elastic coordinate are studied. A high gain feedback controller is designed for the aeroservoelastic model, and closed-loop analysis is conducted based on this model. The simulations show the influence of the elastic motion on the aerodynamic forces and moments.

This paper develops a control method for an air-breathing hypersonic vehicle with variable geometry inlet (AHV-VGI). For the AHV-VGI, a movable translating cowl is used to track the shock on lip conditions to capture enough air mass flow, which can ensure a more powerful thrust. Compared with traditional air-breathing hypersonic vehicle with fixed geometry inlet (AHV-FGI),this AHV-VGI extends the velocity range, which is favorable to the acceleration and maneuvering flight. However, the VG causes the unknown changes of the aerodynamic forces, moment and the thrust in the meanwhile. Therefore, we firstly establish a longitudinal dynamic for AHV-VGI, which includes the uncertain changes induced by VGI. A conception of the optimal elongationdistance of translating cowl is introduced, and its estimated value is obtained by curve fitted approximation. And then, the controlprocess for AHV-VGI is divided into two subsystems. For each subsystem, a sliding mode controller is designed, and interval type-2 fuzzy logic systems (FLSs) are adopted to approximate nonlinear parts including the uncertain changes induced by VGI. Furthermore, uniformly stability of the whole system is proved by Lyapunov approach. Finally, simulation results demonstrate thatAHV have a better control performance under the condition of VGI compared to the FGI.

In this paper, a robust trajectory tracking control for the underactuated quadrotor unmanned aerial vehicle (UAV) based on extended state observer(ESO) is proposed. In the presence of external disturbances and inner model parameter uncertainty, it guaranteed that the position and yaw angle could track the reference signal speedily and smoothly. According to the strict feedback architecture of the system, the backstepping control is adapted to design the controller. For avoiding the complex calculation, the interference is observed by ESO and the algorithm composites the interference in the control law. For avoiding the problem of explosion of terms in backstepping control and improving the robust, the dynamic surface control method and the sliding mode surface are applied to design the controller. The closed-loop control system is analyzed and the sufficient stability conditions are derived based on Lyapunov theory. Numerical examples are presented to verify the proposed method.

This paper investigates the attitude control problem of re-entry Reuseable Launch Vehicles (RLV) with unknown nonlinear dynamics. An adaptive RBF neural network is employed to approximate the unknown functions, and neural network structure with a neural controller and a saturation compensator are designed to reduce the effects of the approximation error. In addition, to optimize the initial parameters of adaptive RBF neural network, a mind evolutionary algorithm is adopted. Furthermore, the adaptive laws of the control system are derived based on Lyapunov stability theorem so that the asymptotical stability of the system can be guaranteed. Finally, Numerical examples are presented to verify the effectiveness of proposed method.

Objective Learning the structures and unknown correlations of a motor imagery electroencephalogram (MI-EEG) signal is important for its classification. It is also a major challenge to obtain good classification accuracy from the increased number of classes and increased variability from different people. In this study, a four-class MI task is investigated. Approach An end-to-end novel hybrid deep learning scheme is developed to decode the MI task from EEG data. The proposed algorithm consists of two parts: a. A one-versus-rest filter bank common spatial pattern is adopted to preprocess and pre-extract the features of the four-class MI signal. b. A hybrid deep network based on the convolutional neural network and long-term short-term memory network is proposed to extract and learn the spatial and temporal features of the MI signal simultaneously. Main results The main contribution of this paper is to propose a hybrid deep network framework to improve the classification accuracy of the four-class MI-EEG signal. The hybrid deep network is a subject-independent shared neural network, which means it can be trained by using the training data from all subjects to form one model. Significance The classification performance obtained by the proposed algorithm on brain-computer interface (BCI) competition IV dataset 2a in terms of accuracy is 83% and Cohen's kappa value is 0.80. Finally, the shared hybrid deep network is evaluated by every subject respectively, and the experimental results illustrate that the shared neural network has satisfactory accuracy. Thus, the proposed algorithm could be of great interest for real-life BCIs.

The variable-geometry scramjet inlet can improve the aerodynamic performance of hypersonic vehicle by moving the movable lip to capture more airflow, which causes the uncertainties of model structures and parameters in the meanwhile. This paper presents a multi-model switching control system to tackle the strong nonlinearity and interaction characteristics. According to the working conditions, we build several nonlinear aerodynamic models with different lengths of the movable lip. For each model, a specific neural network controller has been adopted. By constructing the common Lyapunov function, it is proved that all signals of the closed-looped system are uniformly ultimatedly bounded by the continuous controller. Numerical simulations are presented to verify the feasibility of the proposed control approach.

The problem of adaptive multi-objective optimization (AMOO) has received extensive attention due to its practical significance. An important issue in optimizing a multi-objective system is adjusting the weighting coefficients of multiple objectives so as to keep track of various conditions. In this paper, a feedback structure for AMOO is designed. Moreover, the reinforcement learning combined with hidden biasing information is applied to online tuning weighting coefficients of objective functions. Finally, the proposed approach is applied to the optimization design problem of an elevator group control system. Simulation results show that AMOO has the best average performance at up-peak traffic profile, and its average waiting time reaches 22 s. AMOO is suitable for various traffic patterns, and it is also superior to the majority of algorithms at down-peak traffic profile.

An iterative identification and control design method based on υ-gap is given to ensure the stability of closed-loop system and control performance improvement. The whole iterative procedure includes three parts: the optimal excitation signals design, the uncertainty model set identification and the stable controller design. Firstly the worst case υ-gap is used as the criterion of the optimal excitation signals design, and the design is performed via the power spectrum optimization. And then, an uncertainty model set is attained by system identification on the basis of the measure signals. The controller is designed to ensure the stability of closed-loop system and the closed-loop performance improvement. Simulation result shows that the proposed method has good convergence and closed-loop control performance.