Loading...

Table of Content

    28 October 2020, Volume 25 Issue 5 Previous Issue    Next Issue

    For Selected: Toggle Thumbnails
    Multi-Objective Optimal Feedback Controls for Under-Actuated Dynamical System
    QIN Zhichang, XIN Ying, SUN Jianqiao
    2020, 25 (5):  545-552.  doi: 10.1007/s12204-020-2211-2
    Abstract ( 435 )   PDF (1584KB) ( 15 )  
     This paper presents a study of optimal control design for a single-inverted pendulum (SIP) system with
    the multi-objective particle swarm optimization (MOPSO) algorithm. The proportional derivative (PD) control
    algorithm is utilized to control the system. Since the SIP system is nonlinear and the output (the pendulum angle)
    cannot be directly controlled (it is under-actuated), the PD control gains are not tuned with classical approaches.
    In this work, the MOPSO method is used to obtain the best PD gains. The use of multi-objective optimization
    algorithm allows the control design of the system without the need of linearization, which is not provided by
    using classical methods. The multi-objective optimal control design of the nonlinear system involves four design
    parameters (PD gains) and six objective functions (time-domain performance indices). The Hausdorff distances of
    consecutive Pareto sets, obtained in the MOPSO iterations, are computed to check the convergence of the MOPSO
    algorithm. The MOPSO algorithm finds the Pareto set and the Pareto front efficiently. Numerical simulations
    and experiments of the rotary inverted pendulum system are done to verify this design technique. Numerical and
    experimental results show that the multi-objective optimal controls offer a wide range of choices including the
    ones that have comparable performance to the linear quadratic regulator (LQR) control.
    References | Related Articles | Metrics
    Identification and Control of Flexible Joint Robot Using Multi-Time-Scale Neural Network
    ZHENG Dongdong, LI Pengcheng, XIE Wenfang, LI Dan
    2020, 25 (5):  553-560.  doi: 10.1007/s12204-020-2210-3
    Abstract ( 319 )   PDF (391KB) ( 17 )  
    In this paper, a new identification and control scheme for the flexible joint robotic manipulator is
    proposed. Firstly, by defining some new state variables, the commonly used dynamic equations of the flexible joint
    robotic manipulators are transformed into the standard form of a singularly perturbed model. Subsequently, an
    optimal bounded ellipsoid algorithm based identification scheme using multi-time-scale neural network is proposed
    to identify the unknown system dynamic equations. Lastly, by using the singular perturbation theory, an indirect
    adaptive controller based on the identified model is proposed to control the system such that the joint angles can
    track the given reference signals. The closed-loop stability of the whole system is proved, and the effectiveness of
    the proposed schemes is verified by simulations.
    References | Related Articles | Metrics
    Missile-Target Situation Assessment Model Based on Reinforcement Learning
    ZHANG Yun, Lü Runyan, CAI Yunze
    2020, 25 (5):  561-568.  doi: 10.1007/s12204-020-2226-8
    Abstract ( 455 )   PDF (724KB) ( 15 )  
    In situation assessment (SA) of missile versus target fighter, the traditional SA models generally
    have the characteristics of strong subjectivity and poor dynamic adaptability. This paper considers SA as an
    expectation of future returns and establishes a missile-target simulation battle model. The actor-critic (AC)
    algorithm in reinforcement learning (RL) is used to train the evaluation network, and a missile-target SA model
    is established in simulation battle training. Simulation and comparative experiments show that the model can
    effectively estimate the expected effect of missile attack under the current situation, and it provides an effective
    basis for missile attack decision.
    References | Related Articles | Metrics
    Judging the Normativity of PAF Based on TFN and NAN
    LI Zhiqiang, BAO Jinsong, LIU Tianyuan, WANG Jiacheng
    2020, 25 (5):  569-577.  doi: 10.1007/s12204-020-2177-0
    Abstract ( 291 )   PDF (930KB) ( 7 )  
    The normativity of workers’ actions during producing has a great impact on the quality of the products
    and the safety of the operation process. Previous studies mainly focused on the normativity of each single producing
    action instead of considering the normativity of continuous producing actions, which is defined as producing action
    flow (PAF) in this paper, during operation process. For this issue, a normativity judging method based on two-
    LSTM fusion network (TFN) and normativity-aware attention network (NAN) is proposed. First, TFN is designed
    to detect and recognize the producing actions based on skeleton sequences of a worker during complete operation
    process, and PAF data in sequential form are obtained. Then, NAN is built to allocate different levels of attention
    to each producing action within the sequence of PAF, and by this means, an efficient normativity judging is
    conducted. The combustor surface cleaning (CSC) process of rocket engine is taken as the experimental case, and
    the CSC-Action2D dataset is established for evaluation. Experiment results show the high performance of TFN
    and NAN, demonstrating the effectiveness of the proposed method for PAF normativity judging.
    References | Related Articles | Metrics
    Assembly Information Model Based on Knowledge Graph
    CHEN Zhiyu, BAO Jinsong, ZHENG Xiaohu, LIU Tianyuan
    2020, 25 (5):  578-588.  doi: 10.1007/s12204-020-2179-y
    Abstract ( 365 )   PDF (1606KB) ( 7 )  
    There are heterogeneous problems between the CAD model and the assembly process document. In
    the planning stage of assembly process, these heterogeneous problems can decrease the efficiency of information
    interaction. Based on knowledge graph, this paper proposes an assembly information model (KGAM) to integrate
    geometric information from CAD model, non-geometric information and semantic information from assembly
    process document. KGAM describes the integrated assembly process information as a knowledge graph in the
    form of “entity-relationship-entity” and “entity-attribute-value”, which can improve the efficiency of information
    interaction. Taking the trial assembly stage of a certain type of aero-engine compressor rotor component as an
    example, KGAM is used to get its assembly process knowledge graph. The trial data show the query and update
    rate of assembly attribute information is improved by more than once. And the query and update rate of assembly
    semantic information is improved by more than twice. In conclusion, KGAM can solve the heterogeneous problems
    between the CAD model and the assembly process document and improve the information interaction efficiency.
    References | Related Articles | Metrics
    Micro-Expression Recognition Algorithm Based on Information Entropy Feature
    WU Jin, MIN Yu, YANG Xiaodie, MA Simin
    2020, 25 (5):  589-599.  doi: 10.1007/s12204-020-2219-7
    Abstract ( 270 )   PDF (3121KB) ( 8 )  
    The intensity of the micro-expression is weak, although the directional low frequency components in
    the image are preserved by many algorithms, the extracted micro-expression feature information is not sufficient
    to accurately represent its sequences. In order to improve the accuracy of micro-expression recognition, first, each
    frame image is extracted from its sequences, and the image frame is pre-processed by using gray normalization,
    size normalization, and two-dimensional principal component analysis (2DPCA); then, the optical flow method
    is used to extract the motion characteristics of the reduced-dimensional image, the information entropy value
    of the optical flow characteristic image is calculated by the information entropy principle, and the information
    entropy value is analyzed to obtain the eigenvalue. Therefore, more micro-expression feature information is
    extracted, including more important information, which can further improve the accuracy of micro-expression
    classification and recognition; finally, the feature images are classified by using the support vector machine (SVM).
    The experimental results show that the micro-expression feature image obtained by the information entropy
    statistics can effectively improve the accuracy of micro-expression recognition.
    References | Related Articles | Metrics
    Low-Cost Approach for Improving Video Transmission Efficiency in WVSN
    LIU Min, DENG Bin, TANG Ying, WU Minghu, WANG Juan
    2020, 25 (5):  600-605.  doi: 10.1007/s12204-020-2202-3
    Abstract ( 279 )   PDF (641KB) ( 6 )  
    The wireless visual sensor network (WVSN) as a new emerged intelligent visual system, has been
    applied in many video monitoring sites. However, there is still great challenge because of the limited wireless
    network bandwidth. To resolve the problem, we propose a real-time dynamic texture approach which can detect
    and reduce the temporal redundancy during many successive image frames. Firstly, an adaptively learning background
    model is improved to discover successive similar image frames from the inputting video sequence. Then,
    the dynamic texture model based on the singular value decomposition is adopted to distinguish foreground and
    background element dynamics. Furthermore, a background discarding strategy based on visual motion coherence
    is proposed to determine whether each image frame is streamed or not. To evaluate the trade-off performance
    of the proposed method, it is tested on the CDW-2014 dataset, which can accurately detect the first foreground
    frame when the moving objects of interest appear in the field of view in the most tested dynamic scenes, and
    the misdetection rate of the undetected foreground frames is near to zero. Compared to the original stream, it
    can reduce the occupied bandwidth a lot and its computational cost is relatively lower than the state-of-the-art
    methods.
    References | Related Articles | Metrics
    Simulation Research on Operation of Union Purchase System in Navigation Simulator
    FAN hang, YANG Shenhua, SUO Yongfeng, ZHENG Minjie
    2020, 25 (5):  606-614.  doi: 10.1007/s12204-020-2193-0
    Abstract ( 306 )   PDF (2104KB) ( 7 )  
    Navigation is the only way to develop and utilize marine resources, while the promotion of seafarers’
    quality is the basic force of navigation, so navigation simulator plays an important role in modern navigation
    education. The simulation research on the operation of the union purchase is important to improve the special
    operation training of the actual cargo handling of the union purchase. On the basis of the Cartesian coordinate
    system transformation algorithm, the algorithm model of the union purchase operation is constructed. On the
    basis of three-dimensional (3D) rendering engine technology of open scene graph (OSG), the algorithm model
    of finding the space coordinates of the cargo point is established. The model of catenary equation is used to
    optimize the scene appearance of the cargo wire. By combination of QT channel signal mechanism and OSG, the
    simulation interaction of the union purchase operating system is realized. By acquiring the 3D coordinate values
    of each point, we fit the trajectories of each point in the operation and compare the trajectories. The results
    show that the model has high interactivity and small error. The comparison of the states of the cargo wire before
    and after optimization shows that the optimized wire is more realistic and the high fidelity meets the needs of
    operational training and simulation systems.
    References | Related Articles | Metrics
    BFM: A Bus-Like Data Feedback Mechanism Between Graphics Processor and Host CPU
    DENG Junyong, JIANG Lin
    2020, 25 (5):  615-622.  doi: 10.1007/s12204-020-2221-0
    Abstract ( 296 )   PDF (575KB) ( 7 )  
    Graphics processors have received an increasing attention with the growing demand for gaming, video
    streaming, and many other applications. During the graphics rendering with OpenGL, host CPU needs the runtime
    attributes to move on to the next procedure of rendering, which covers almost all the function units of
    graphics pipeline. Current methods suffer from the memory capacity issues to hold the variables or huge amount
    of data passing paths which can cause congestion on the interface between graphics processor and host CPU. This
    paper refers to the operation principle of commuting bus, and proposes a bus-like data feedback mechanism (BFM)
    to traverse all the pipeline stages and collect the run-time status data or execution error of graphics rendering,
    then send them back to the host CPU. BFM can work in parallel with the graphics rendering logic. This method
    can complete the data feedback task easily with only 0.6% increase of resource utilization and has no negative
    impact on performance, which also obtains 1.3 times speed enhancement compared with a traditional approach.
    References | Related Articles | Metrics
    Numerical Computation of a Mixed-Integer Optimal Control Problem Based on Quantum Annealing
    LIU Zhe, LI Shurong, GE Yulei
    2020, 25 (5):  623-629.  doi: 10.1007/s12204-020-2220-1
    Abstract ( 346 )   PDF (211KB) ( 7 )  
    It is extremely challenging to solve the mixed-integer optimal control problems (MIOCPs) due to
    the complex computation in solving the integer decision variables. This paper presents a new method based
    on quantum annealing (QA) to solve MIOCP. The QA is a metaheuristic which applies quantum tunneling in
    the annealing process. It has a faster convergence speed in optimal-searching and is less likely to run into local
    minima. Hence, QA is applied to deal with this kind of optimization problems. First, MIOCP is transformed into
    a mixed-integer nonlinear programming (MINLP). Then, a method based on QA is adopted to solve the MINLP
    and acquire the optimal solution. At last, two benchmark examples including Lotka-Volterra type fishing problem
    and distillation column are presented and solved. The effectiveness of the methodology is verified by the acquired
    optimal schemes.
    References | Related Articles | Metrics
    Fault Reconstruction for Lipschitz Nonlinear Systems Using Higher Terminal Sliding Mode Observer
    DAI Cong, LIU Yongzhi, SUN Haoshui
    2020, 25 (5):  630-638.  doi: 10.1007/s12204-020-2196-x
    Abstract ( 269 )   PDF (443KB) ( 13 )  
    This paper considers the design of an adaptive second order terminal observer for robust fault reconstruction
    of nonlinear Lipschitz systems with unknown upper bound of derivative fault. Firstly, a linear
    transforming matrix is introduced, which transforms the system into two subsystems, and thus to reduce the
    dimension of the system. One of the subsystem is affected by fault and disturbances, while the other is free,
    which simplifies the design of observer. Then, the design method of the observer gain matrix is transformed into
    a convex optimization problem under linear matrix inequalities (LMIs). A second order non-singular terminal
    sliding mode observer is designed for the transformed system to realize the accurate estimation of state and fault.
    Considering the unknown upper bound of derivative fault, an adaptive algorithm is designed in the equivalent
    output error injection signal to ensure the sliding mode motion reach the sliding surface within limited time.
    Finally, an example demonstrates the effectiveness of the proposed method in the paper.
    References | Related Articles | Metrics
    5D Hyper-Chaotic System with Multiple Types of Equilibrium Points
    XU Changbiao, WU Xia, HE Yinghui, MO Yunhui
    2020, 25 (5):  639-649.  doi: 10.1007/s12204-020-2224-x
    Abstract ( 255 )   PDF (1544KB) ( 10 )  
     A chaotic system with various equilibrium types has rich dynamic behaviors. Its state can switch
    flexibly among different families of attractors, which is beneficial to the practical applications. So it has been widely
    concerned in recent years. In this paper, a new 5D hyper-chaotic system is proposed. The important characteristic
    of the system is that it may have multiple types of equilibrium points by changing system parameters, namely,
    linear equilibrium point, no equilibrium point, non-hyperbolic unstable equilibrium point and stable hyperbolictype
    equilibrium point. Furthermore, there are hyper-chaotic phenomena and multi-stability about the coexistence
    of multiple chaotic attractors and the coexistence of hyper-chaotic attractors and chaotic attractors in the system.
    In addition, the system’s complexity is analyzed. It is found that the complexity is close to 1 in the hyper-chaotic
    state and a pseudo-random sequence generated by the system passes all the statistical tests. Finally, an analog
    circuit of the system is designed and simulated.
    References | Related Articles | Metrics
    Low Data Overlap Rate Graph-Based SLAM with Distributed Submap Strategy
    XIANG Jiawei, ZHANG Jinyi, WANG Bin, MA Yongbin
    2020, 25 (5):  650-658.  doi: 10.1007/s12204-020-2201-4
    Abstract ( 281 )   PDF (1218KB) ( 8 )  
     Simultaneous localization and mapping (SLAM) is widely used in many robot applications to acquire
    the unknown environment’s map and the robot’s location. Graph-based SLAM is demonstrated to be effective in
    large-scale scenarios, and it intuitively performs the SLAM as a pose graph. But because of the high data overlap
    rate, traditional graph-based SLAM is not efficient in some respects, such as real time performance and memory
    usage. To reduce data overlap rate, a graph-based SLAM with distributed submap strategy (DSS) is presented.
    In its front-end, submap based scan matching is processed and loop closing detection is conducted. Moreover in
    its back-end, pose graph is updated for global optimization and submap merging. From a series of experiments, it
    is demonstrated that graph-based SLAM with DSS reduces 51.79% data overlap rate, decreases 39.70% runtime
    and 24.60% memory usage. The advantages over other low overlap rate method is also proved in runtime, memory
    usage, accuracy and robustness performance.
    References | Related Articles | Metrics
    Data-Driven Fault Detection of Three-Tank System Applying MWAT-ICA
    LIU Mingguang, LIAO Yaxuan, LI Xiangshun
    2020, 25 (5):  659-664.  doi: 10.1007/s12204-020-2227-7
    Abstract ( 335 )   PDF (212KB) ( 8 )  
    In order to improve monitoring performance of dynamic process, a moving window independent component
    analysis method with adaptive threshold (MWAT-ICA) is proposed. On-line fault detection can be realized
    by applying moving windows technique, as well as false alarm caused by fluctuation of data can be effectively
    avoided by adaptive threshold. The efficiency of the proposed approach is demonstrated with a three-tank system.
    The results show that the MWAT-ICA can not only detect the fault quickly, but also has a high fault detection
    rate and no false alarm rate under the transient behaviors of the three-water tank and the normal operation
    process. These results demonstrate the effectiveness of the method for fault detection on the three-tank system.
    References | Related Articles | Metrics
    Dual Sum-Product Networks Autoencoder for Multi-Label Classification
    WANG Shengsheng, ZHANG Hang, CHEN Juan
    2020, 25 (5):  665-673.  doi: 10.1007/s12204-020-2204-1
    Abstract ( 262 )   PDF (218KB) ( 6 )  
    Sum-product networks (SPNs) are an expressive deep probabilistic architecture with solid theoretical
    foundations, which allows tractable and exact inference. SPNs always act as black-box inference machine in many
    artificial intelligence tasks. Due to their recursive definition, SPNs can also be naturally employed as hierarchical
    feature extractors. Recently, SPNs have been successfully employed as autoencoder framework in representation
    learning. However, SPNs autoencoder ignores the model structural duality and trains the models separately and
    independently. In this work, we propose a Dual-SPNs autoencoder which designs two SPNs autoencoders to
    compose as a dual form. This approach trains the models simultaneously, and explicitly exploits the structural
    duality between them to enhance the training process. Experimental results on several multilabel classification
    problems demonstrate that Dual-SPNs autoencoder is very competitive against with state-of-the-art autoencoder
    architectures.
    References | Related Articles | Metrics
    Finite-Time Stability and Stabilization of Discrete-Time Switching Markov Jump Linear System
    JIN Yunyun, SONG Yang, LIU Yongzhuang, HOU Weiyan
    2020, 25 (5):  674-680.  doi: 10.1007/s12204-020-2205-0
    Abstract ( 292 )   PDF (199KB) ( 8 )  
    Switching Markov jump linear system (SMJLS), a special hybrid system, has attracted a lot of studies
    recently. SMJLS is governed by stochastic and deterministic commutations. This paper focuses on the switching
    strategy which stabilizes the SMJLS in a finite time interval in order to further expand the existing results
    and investigate new aspects of such systems. Several sufficient conditions for finite-time stability of discrete-time
    SMJLS are provided, and the numerical problems in these sufficient conditions are solved by solving linear matrix
    inequalities (LMIs). Finally, numerical examples are given to show the feasibility and effectiveness of the results.
    References | Related Articles | Metrics