Search

Multiple object tracking plays a crucial role in intelligent transportation systems. Due to the varying size, fast motion, and occlusion of traffic objects, multiple object tracking in complex traffic scenes is prone to low tracking accuracy and tracking fragmentation. To solve these problems, numerous tracking methods have been proposed based on object discrimination feature. However, these methods neglect the guiding prior role of historical tracking information for detection, making them not applicable to more complex scenes, such as extremely small, fast-moving, and severely obscured objects. In this paper, we propose an Interactively Motion-assisted Network (IMANet) for multiple object tracking in complex traffic scenes. First, to capture the object motion patterns from historical frames, an object motion modeling module considering camera movement is proposed, which particularly has prominent advantages on videos obtained by moving cameras. Next, a multi-scale fusion detection and embedding module is designed to incorporate historical motion information, thereby improving detection performance. Finally, multiple object tracking can be achieved by associating objects detected in different frames based on detection and embedding results. The proposed method combines the detection and tracking in an interactive way, where detection performance is facilitated using historical information provided by the tracking module, and better detection in turn enhances the tracking. Several real-world traffic examples are used to illustrate the performance of the proposed method in both detecting and tracking traffic objects. The results demonstrate that the proposed method outperforms the state-of-the-art methods, especially in complex surveillance videos with varying sizes and occlusions.

In the past decades, radar-based hand gesture recognition (HGR) has gained increased attention in several applications involving contactless human-computer interaction (HCI). In this article, we propose both macro hand and micro finger gesture recognitions using an interferometric multi-input multi-output (MIMO) radar. The two principal signal processing modes of the radar are the conventional MIMO mode for initial positioning and the transmit interferometric mode for trajectory tracking. To achieve high precision sensing, a novel temporal and spatial interferometry is applied to acquire the subtle range and angular displacements of hand/finger motions, in lieu of the commonly used micro-Doppler (mD) spectrogram. Additionally, a ResNet50 convolution neural network (CNN) trained with 3-D space-time coordinates is used to provide HGR robustness against similar drawings. Simulation and experimental results show that the proposed interferometric MIMO radar performs rather well in sensing and tracking hand movements, achieving a recognition accuracy of 96.64%–96.33% for macro hand and micro finger gestures, respectively, and outperforming existing HGR methods.

Global navigation satellite system (GNSS) interference severely affects the quality of automatic dependent surveillance-broadcast (ADS-B) data, thereby jeopardizing aviation safety. Therefore, this article proposes an attention-based machine learning methodology for detecting and localizing GNSS interference sources. By exploring the spatio-temporal relationship between the interfered ADS-B data and the GNSS interference source, the interference source can be detected and located. Initially, we use the logistic regression algorithm to approximately detect and locate interference in an area. Subsequently, we propose an attention mechanism convolutional network (AMCN) to accurately localize interference sources. The proposed AMCN comprises two main blocks: A convolutional network that captures local features of individual aircraft and an attention network that captures the overall features of all associated aircraft. Within the attention network, an improved position embedding method maps sample sequences to their actual spatial locations. We demonstrate the effectiveness of our approach by achieving significant improvements over the state-of-the-art on actual aviation data. The proposed approach has the potential to effectively detect and locate GNSS interference sources, thereby reducing the security risk in civil aviation.

Due to the significance of air traffic delay propagation in the operational robustness of air transport systems, it has recently received increased attention from both industry and academia. The purpose of delay propagation analysis is to capture the dynamic traveling trajectories of root delays in air traffic networks. Few studies have investigated delay propagation while taking consideration of temporal context information, leading to incomplete traveling trajectories of root delays. In this paper, a novel systematic framework based on temporal causal inference is proposed to model delay propagation in airport networks. More specifically, considering the lagged nature caused by the transition time in delay propagation, a fully-connected delay propagation network based on transfer entropy is developed. Additionally, a causal biased random walk is embedded to explore the temporal cascading effect of root delays in air transport systems and to generate delay propagation trees for each airport. Extensive experiments on a real-world dataset indicate that our framework is able to effectively model the dynamic evolution of air traffic delays in an airport network with acceptable performance. This paper also presents a real-world case study of the Chinese airport network, which reveals that the traveling trajectories of air traffic delays differ significantly at various daily delay levels and airport throughput.

To provide seamless coverage during all flight phases, aeronautical communications systems (ACS) have to integrate space-based, air-based, as well as ground-based platforms to formulate aviation-oriented space-air-ground integrated networks (SAGINs). In continental areas, L-band aeronautical broadband communications (ABC) are gaining popularity for supporting air traffic management (ATM) modernization. However, L-band ABC faces the challenges of spectrum congestion and severe interference due to the legacy systems. To circumvent these, we propose a novel multiple-antenna aided L-band ABC paradigm to tackle the key issues of reliable and high-rate air-to-ground (A2G) transmissions. Specifically, we first introduce the development roadmap of the ABC. Further-more, we discuss the peculiarities of the L-band ABC propagation environment and the distinctive challenges of the associated multiple-antenna techniques. To overcome these challenges, we propose an advanced multiple-antenna assisted L-band ABC paradigm from the perspective of channel estimation, reliable transmission, and multiple access. Finally, we shed light on the compelling research directions of the aviation component of SAGINs.

How to establish control design methods for learning systems that are subjected to the stochastic uncertainties becomes a topic of practical importance in the control field. This paper deals with the stochastic iterative learning control (ILC) problems for linear time-varying systems subject to measurement noises and changing durations. The lengths of the changing durations are modeled as a Markov chain with finite states. By exploring the essential trackability problem, we develop a trackability-based design and analysis framework for the stochastic ILC systems. Besides, we design three trackability-based P-type stochastic ILC updating laws to achieve the tracking tasks, with some simple gain matrix selection conditions. Further, the convergence results for the stochastic ILC systems in both the almost-sure and mean-square senses are developed. Two illustrative examples are included to demonstrate the effectiveness of the trackability-based stochastic ILC results.

This article focuses on exploring a class of Lyapunov analysis approaches to deal with the convergence problems on second-order signed networks (SOSNs) under arbitrary strongly connected directed topologies. A new class of Laplacian potentials is first proposed for SOSNs by exploiting the properties of Laplacian matrices. Then the relation between convergence behaviors of all agents and those of Laplacian potentials can be disclosed, which makes it possible to solve the convergence issues of SOSNs from the viewpoint of the Lyapunov stability theory even though the weight-balanced condition is not satisfied. Furthermore, the proposed Laplacian potential can be leveraged to deal with the convergence problems of distributed averaging for SOSNs. It is shown that the signed-average consensus objective is reached under the structurally balanced signed digraphs by designing a distributed control protocol according to the proposed Laplacian potential. Additionally, simulation examples are given to demonstrate the validity of our potential-based distributed control results.