Adaptive traffic signal control plays a significant role in the construction of smart cities. This task is challenging because of many essential factors, such as cooperation among neighboring intersections and dynamic traffic scenarios. First, to fac
ilitate cooperation of traffic signals, existing work adopts graph neural networks to incorporate the temporal and spatial influences of the surrounding intersections into the target intersection, where spatial-temporal information is used separately. However, one drawback of these methods is that the spatial-temporal correlations are not adequately exploited to obtain a better control scheme. Second, in a dynamic traffic environment, the historical state of the intersection is also critical for predicting future signal switching. Previous work mainly solves this problem using the current intersections state, neglecting the fact that traffic flow is continuously changing both spatially and temporally and does not handle the historical state. In this paper, we propose a novel neural network framework named DynSTGAT, which integrates dynamic historical state into a new spatial-temporal graph attention network to address the above two problems. More specifically, our DynSTGAT model employs a novel multi-head graph attention mechanism, which aims to adequately exploit the joint relations of spatial-temporal information. Then, to efficiently utilize the historical state information of the intersection, we design a sequence model with the temporal convolutional network (TCN) to capture the historical information and further merge it with the spatial information to improve its performance. Extensive experiments conducted in the multi-intersection scenario on synthetic data and real-world data confirm that our method can achieve superior performance in travel time and throughput against the state-of-the-art methods.
Reinforcement learning (RL) is always the preferred embodiment to construct the control strategy of complex tasks, like asymmetric assembly tasks. However, the convergence speed of reinforcement learning severely restricts its practical application.
In this paper, the convergence is first accelerated by combining RL and compliance control. Then a completely innovative progressive extension of action dimension (PEAD) mechanism is proposed to optimize the convergence of RL algorithms. The PEAD method is verified in DDPG and PPO. The results demonstrate the PEAD method will enhance the data-efficiency and time-efficiency of RL algorithms as well as increase the stable reward, which provides more potential for the application of RL.
This paper aims at solving mass precise peg-in-hole assembly. First, a feature space and a response space are constructed according to the relative pose and equivalent forces and moments. Then the contact states are segmented in the feature space and
the segmentation boundaries are mapped into the response space. Further, a feature-based compliance control (FBCC) algorithm is proposed based on boundary mapping. In the FBCC algorithm, a direction matrix is designed to execute accurate adjustment and an integrator is applied to eliminate the residual responses. Finally, the simulations and experiments demonstrate the superiority, robustness, and generalization ability of the FBCC.
This paper defines a security injection region (SIR) to guarantee reliable operation of water distribution systems (WDS) under extreme conditions. The model of WDSs is highly nonlinear and nonconvex. Understanding the accurate SIRs of WDSs involves t
he analysis of nonlinear constraints, which is computationally expensive. To reduce the computational burden, this paper first investigates the convexity of the SIR of WDSs under certain conditions. Then, an algorithm based on a monotone inner polytope sequence is proposed to effectively and accurately determine these SIRs. The proposed algorithm estimates a sequence of inner polytopes that converge to the whole convex region. Each polytope adds a new area to the SIR. The algorithm is validated on two different WDSs, and the conclusion is drawn. The computational study shows this method is applicable and fast for both systems.
Reconfigurable intelligent surface (RIS) is an emerging technique employing metasurface to reflect the signal from the source node to the destination node without consuming any energy. Not only the spectral efficiency but also the energy efficiency c
an be improved through RIS. Essentially, RIS can be considered as a passive relay between the source and destination node. On the other hand, a relay node in a traditional relay network has to be active, which indicates that it will consume energy when it is relaying the signal or information between the source and destination nodes. In this paper, we compare the performances between RIS and active relay for a general multiple-input multiple-output (MIMO) system. To make the comparison fair and comprehensive, both the performances of RIS and active relay are optimized with best-effort. In terms of the RIS, transmit beamforming and reflecting coefficient at the RIS are jointly optimized so as to maximize the end-to-end throughput. Although the optimization problem is non-convex, it is transformed equivalently to a weighted mean-square error (MSE) minimization problem and an alternating optimization problem is proposed, which can ensure the convergence to a stationary point. In terms of active relay, both half duplex relay (HDR) and full duplex relay (FDR) are considered. End-to-end throughput is maximized via an alternating optimization method. Numerical results are presented to demonstrate the effectiveness of the proposed algorithm. Finally, comparisons between RIS and relays are investigated from the perspective of system model, performance, deployment and controlling method.
A compressive sensing based circular polarization snapshot spectral imaging system is proposed in this paper to acquire two-dimensional spatial, one-dimensional circular polarization (the right and left circular polarization), and one-dimensional spe
ctral information, simultaneously. Using snapshot can collect the entire four-dimensional datacube in a single integration period. The dispersion prism in the coded aperture snapshot spectral imager is replaced by the combination of an Amici prism and a Wollaston prism to implement the spectral shifting along two orthogonal directions, which greatly improves the spectral resolution of the image. The right and left circular polarization components of objects are extracted by the assemble with an achromatic quarter wave-plate and a Wollaston prism. The encoding and reconstruction are illustrated comprehensively. The feasibility is verified by the simulation. It provides us an alternative approach for circular polarization spectral imaging such as defogging, underwater imaging, and so on.
In recent years, semi-supervised multi-view nonnegative matrix factorization (MVNMF) algorithms have achieved promising performances for multi-view clustering. While most of semi-supervised MVNMFs have failed to effectively consider discriminative in
formation among clusters and feature alignment from multiple views simultaneously. In this paper, a novel Discriminatively Constrained Semi-Supervised Multi-View Nonnegative Matrix Factorization (DCS^2MVNMF) is proposed. Specifically, a discriminative weighting matrix is introduced for the auxiliary matrix of each view, which enhances the inter-class distinction. Meanwhile, a new graph regularization is constructed with the label and geometrical information. In addition, we design a new feature scale normalization strategy to align the multiple views and complete the corresponding iterative optimization schemes. Extensive experiments conducted on several real world multi-view datasets have demonstrated the effectiveness of the proposed method.
Voltage instability is one of the main causes of power system blackouts. Emerging technologies such as renewable energy integration, distributed energy resources and demand responses may introduce significant uncertainties in analyzing of system-wide
voltage stability. This paper starts with summarizing different known voltage instability mechanisms, and then focuses on a class of voltage instability which is induced by the singular surface of the algebraic manifold. We argue and demonstrate that this class can include both dynamic and static voltage instabilities. To determine the minimum distance to the point of voltage collapse, a new formulation is proposed on the algebraic manifold. This formulation is further converted into an optimal control framework for identifying the path with minimum distance on the manifold. Comprehensive numerical studies are conducted on some manifolds of different power system test cases and demonstrate that the proposed method yields candidates for the local shortest paths to the singular surface on the manifold for both the dynamic model and the static model. Simulations show that the proposed method can identify shorter paths on the manifold than the paths associated with the minimum Euclidean distances. Furthermore, the proposed method always locates the right path ending at the correct singular surface which is responsible for the voltage instability; while the Euclidean distance formulation can mistakenly find solutions on the wrong singular surface. A broad range of potential applications using the proposed method are also discussed.
In this paper, we establish various maximal principles and develop the direct moving planes and sliding methods for equations involving the physically interesting (nonlocal) pseudo-relativistic Schr{o}dinger operators $(-Delta+m^{2})^{s}$ with $sin(0
,1)$ and mass $m>0$. As a consequence, we also derive multiple applications of these direct methods. For instance, we prove monotonicity, symmetry and uniqueness results for solutions to various equations involving the operators $(-Delta+m^{2})^{s}$ in bounded domains, epigraph or $mathbb{R}^{N}$, including pseudo-relativistic Schrodinger equations, 3D boson star equations and the equations with De Giorgi type nonlinearities.
Power system cascading failures become more time variant and complex because of the increasing network interconnection and higher renewable energy penetration. High computational cost is the main obstacle for a more frequent online cascading failure
search, which is essential to improve system security. In this work, we show that the complex mechanism of cascading failures can be well captured by training a graph convolutional network (GCN) offline. Subsequently, the search of cascading failures can be significantly accelerated with the aid of the trained GCN model. We link the power network topology with the structure of the GCN, yielding a smaller parameter space to learn the complex mechanism. We further enable the interpretability of the GCN model by a layer-wise relevance propagation (LRP) algorithm. The proposed method is tested on both the IEEE RTS-79 test system and Chinas Henan Province power system. The results show that the GCN guided method can not only accelerate the search of cascading failures, but also reveal the reasons for predicting the potential cascading failures.
