Search

Communication overhead is a key task for the large-scale wireless sensor networks (WSNs) in the cooperative localization problems. This paper is aimed to propose a low-overhead algorithm without sacrificing the performance of localization, which is more practical for energy-saving WSNs. In this paper, the cooperative localization problem is formulated as a variational reference problem on a factor graph (FG). Combining the high performance of particle-based algorithms and the overhead advantage of parametric ones, a parametric variational sum-product algorithm (PVSPA) is proposed for the cooperative localization of WSNs. In prediction operation, to ensure high accuracy, messages referenced inside sensor nodes are particle-based. In correction operation, Gaussian parametric representation is adopted to the observation message and the communication overhead is obviously decreased. Simulation results show that the localization accuracy of PVSPA equally matches the classic particle-based algorithms, while the communication overhead of it is far lower.

Student performance prediction is a fundamental task in online learning systems, which aims to provide students with access to active learning. Generally, student performance prediction is achieved by tracing the evolution of each student's knowledge states via a series of learning activities. Every learning activity record has two types of feature data: student behavior and exercise features. However, most methods use features that are related to exercises, such as correctness and concepts, while other student behavior features are usually ignored. The few studies that have focused on student behavior features through subjective manual selection argue that different student behavior features can be used in an equivalent manner to predict student performance. In this paper, we assume that the integration of student behavior features and exercise features is crucial to improve the precision of prediction, and each feature has a different impact on student performance. Therefore, this paper proposes a novel framework for student performance prediction by making full use of both student behavior features and exercise features and combining the attention mechanism with the knowledge tracing model. Specifically, we first exploit machine learning to capture feature representation automatically. Then, a fusion attention mechanism based on recurrent neural network architecture is used for student performance prediction. Extensive experiments on a real-world dataset show the effectiveness and practicability of our approach. The accuracy of our method is up to 98%, which is superior to previous methods.

This paper investigates model reference control for linear time-varying (LTV) systems. It is shown that the problem can be decomposed into two sub-problems: a state feedback stabilization problem of LTV systems and a feed-forward compensation problem. Firstly, the sufficient condition for the existence of the model reference control is deduced. The condition concerns two coefficient matrices such that two matrix equations are met simultaneously, and the state feedback gain matrix stabilizes LTV systems according to the unified spectral theory. Secondly, generally complete parametrization of the model reference tracking controller including a state feedback controller and a feed-forward compensator is proposed based on the solution to a type of generalized Sylvester matrix equations. Furthermore, a general procedure for solving the model reference control problem is also provided. Finally, a numerical example shows the feasibility and effectiveness of the parametric design approach to the model reference control in LTV systems.

This paper concentrates on the design of intelligent adaptive tracking controller for stochastic switched nonlinear pure-feedback systems with input saturation and non-lower triangular structure. It needs to be emphasized that both the issues of pure-feedback structure and non-differential saturation nonlinearity are involved in the studied system. With the help of the mean-value theorem, a novel intelligent adaptive tracking controller is developed in this work to overcome the difficulty resulted from pure-feedback structure, and the inherent property of Gaussian functions is utilized to handle functions that are unknown and include all state variables. Moreover, through the universal intelligent approximation technology, a novel control strategy is constructed under the framework of backstepping, which guarantees that the tracking error can converge to a small neighborhood near the origin in the sense of mean quartic value and all signals of the nonlinear closed-loop system can be bounded in probability. Eventually, the effectiveness of the presented scheme is further illustrated by the simulation of two practical examples.

In view of the inherent defects of traditional network protocols, unreasonable video content distribution process, content scheduling and routing mechanism need to be optimized in the current edge computing (EC) network. A centralized control EC network architecture of named data networking (NDN) is proposed in this paper. EC nodes are deployed on the edge of the network to provide caching and computing capabilities, and distribute video streams. So that transmission delay is reduced, bandwidth cost is saved, and the quality of experience (QoE) is improved. Firstly, the communication mechanism of NDN is introduced to distribute content and improve the efficiency of content distribution. In the scheme, the content distribution mechanism of video stream is improved in a distributed way. Different bit rate versions of video clips are distributed and buffered on the distributed EC nodes, so as to avoid redundant cache and save storage resources. In addition, the user interest content model is constructed. That is, video is divided into different categories according to content and user content preference model. Finally, with the limited cache and computing resources of EC nodes, the software defined network (SDN) mechanism is introduced for centralized management and control. And it provides a solution for content scheduling problem under resource constraints. In the scheme, the content distribution mode is decided by the effective arrangement of network resources, and the user's request is processed and responded cooperatively to avoid redundant cache and transmission. Based on the selection of forwarding response mode, the distributed characteristics of EC nodes and the routing and forwarding mechanism of NDN are fully utilized to provide users with response. It is shown by the experimental results that the proposed scheme optimizes the load on the network side and the experience of the client to a certain extent.

Lane changing (LC) is the basic maneuver of a variety of lateral control scenes such as overtaking, paralleling, collision avoidance, etc. The surrounding vehicle may drive into the planned path during the ego-vehicle’s LC maneuver, the ego-vehicle could only stop or fail to respond in time. Aiming to tackle this issue, this article proposes a LC strategy with real-time obstacle avoidance function. Firstly, the common LC strategy adopts the single planning and real-time tracking framework, which cannot handle emergent situations during the LC maneuver. This article proposes a new framework of real-time planning and real-time tracking for obstacle avoidance in the process of the LC. Secondly, this article constructs a two-layer judgment condition and control mode matrix for the two obstacle avoidance conditions of current-lane preceding-vehicle lane-change (CPC) and target-lane preceding-vehicle deceleration (TPD). Finally, experiments and simulations demonstrate the effectiveness of the proposed algorithm.

Due to the limited space on the train, the shielded cables were utilized to protect signal cables against influence by the disturbance from power cables which arranged in parallel with the signal cables. To reduce cable coupling, the cables are usually arranged close to train body. The influence of the body of the train should not be ignored. To analyze the coupling of shielded cables in the train, a hybrid method is proposed based on the circuit model. To solve the influence from the reflection of the train body, Green's function deduced from image theory is utilized. The current on the shielding layer is calculated by lumped circuit model, which is set up by admittance parameter matrix. And then, the coupling voltage on the core of shielded cable is given by using BTL function. Finally, a laboratory test case and a field test were investigated. Results of the proposed method were validated by measurement results.

Autonomous underwater vehicles (AUVs) are robots that require path planning to complete missions in different kinds of underwater environments. The goal of path planning is to find a feasible path from the start-point to the target-point in a given environment. In most practical applications, environments have dynamic factors, such as ocean flows and moving obstacles, which make the AUV path planning more challenging. This paper proposes an estimation of distribution algorithm (EDA) based approach, termed as learning fixed-height histogram (LFHH) to solve path planning problems for AUVs in dynamic environment. The LFHH uses a learning transformation strategy (LTS) to improve its accuracy and convergence speed. Besides, a smooth method is employed to accelerate the speed of finding feasible paths. Moreover, a planning window is adopted to help handle dynamic factors. LFHH is tested in both complex 2-D and 3-D environments with time-variant dynamic factors, and experimental results validate the effectiveness of LFHH.

In low-altitude target situation, the multi-path signals cause the amplitude-phase distortion of direct signal from targets and degrade the performance of existing methods. Hence, in this paper, we propose a phase enhancement method for low-angle estimation using supervised deep neural network (DNN) to mitigate the phase distortion, thus to improve direction of arrival (DOA) estimation accuracy. The mapping relationship between the original phase difference distribution of the received signal and desired phase difference distribution is learned by DNN during training. The phase of test data is enhanced by trained DNN, and the enhanced phase is used for DOA estimation. We explain the significance of enhancing phase instead of amplitude by discussing the sensitivity of amplitude and phase on DOA estimation. Moreover, we prove the effectiveness and superiority of the proposed method by simulation experiments. The results demonstrate that the proposed technique has a better performance in terms of estimation error and goodness of fit (GoF) than the physics-driven DOA estimation methods and state-of-the-art methods including feature reversal and the support vector regression (SVR).

Vegetation is an important part of the ecological channel, and flow structure over vegetation in ecological channel is rather complex. The vegetation resistance to flow is affected by many factors, and there is still no general calculation method to express it. Thus, it is necessary to explore the mechanism of flow resistance over vegetation in open channel. In this paper, the rigid cylindrical sticks arranged in the open channel were used to simulate the stems of non-submerged vegetation in order to investigate the effects of vegetation on the flow. Through theoretical analysis and experimental verification, it shows the form drag caused by vegetation is closely related to the vortex volume created by vegetation. In addition, the total resistance of vegetated channel can be determined by two methods based on vegetation density and flow resistance partitioning, respectively, and both of them are verified by the experimental results. What's more, for rigid vegetation, the drag coefficient is closely related to vegetation density, spacing between vegetation patches and the Reynolds number. The empirical formula among them was established through the experimental results.

This paper presents the parametric design approach to the second-order quasi-linear (SQ) systems by using dynamic compensator and multi-objective optimization. Based on the solutions to a type of the second-order generalized Sylvester equations (SGSEs), the generally parametric form of the dynamic compensator is established, the completely parameterized expressions of left and right eigenvector matrices are obtained, and it also provides two groups of arbitrary parameter matrices. With the parametrized method, the closed-loop system is converted into a linear constant one. Simultaneously, it also investigates a novel technique to multi-objective design and optimization. Multiple performance indexes, such as regional pole assignment, low sensitivity, disturbance attenuation, robustness degree, and low gains, are formulated by arbitrary parameters. Based on the above indexes, robustness criteria and low gain criteria can be expressed by a synthetic objective function which includes each performance index weighted. By using the degrees of freedom (DOFs) in arbitrary parameters, a dynamic compensator can be established by solving a multi-objective optimization problem. Finally, the spacecraft rendezvous problem is proposed to verify the feasibility and effectiveness of the parametrized approach.

The flow structure over vegetation in ecological channel is rather complex and the resistance caused by vegetation is affected by many factors, such as discharge, flow depth, vegetation density and patch arrangement types, etc. In order to simplify vegetation characteristics, the rigid cylindrical sticks arranged in the open channel were used to simulate the stems of non-submerged vegetation. The volume of the vortex structure near the vegetation patch can be determined through velocity distribution in vegetated channel and the velocity distribution is measured by Acoustic Doppler Velocity meter. A new formula of equivalent roughness based on the vortex volume near the vegetation patch has been proposed and the form drag can be further determined by the new equivalent roughness. Based on theoretical analysis and experimental results, it shows the vortex volume plays an important role on flow resistance caused by vegetation. The theoretical and experimental results on form drag achieve a good agreement on various experimental conditions. The relation between the relative roughness and relative vortex length was also derived and verified by different experimental conductions. For the dependence of relative roughness on vegetation density, the empirical formula between relative roughness and vegetation density was further proposed and analyzed.