Space-air-ground integrated vehicular networks (SAGIVNs) can provide long-distance communication service with wide coverage for ground users such as vehicles. However, scarce spectrum resources and long-distance transmission result in high path loss as well as latency. As relay stations, unmanned aerial vehicles (UAVs) play key roles in improving transmission quality between space and ground, i.e., satellites and vehicles, whose configuration of resources highly influences network performance. In this paper, we focus on joint resource allocation for UAVs in SAGIVNs. Specifically, the joint optimization problem in user association, trajectory design, and power control is addressed. By considering the trade-off between transmission rate and energy consumption in SAGIVNs, we propose an energy-efficient joint resource allocation approach named ERDM. Firstly, vehicles are assigned to UAVs based on their positions. hen, we formulate the optimization problem to maximize energy efficiency, the sum rate with unit power cost, with consideration of service quality constraints. To reduce the complexity of the problem, we employ multi-agent deep reinforcement learning framework to obtain optimal solutions. By incorporating the behavior of other agents, UAVs learn to allocate resources for themselves and update Q-networks using gained experiences in distinctive observations and feedback rewards. Simulations reveal ERDM outperforms other benchmarks by up to 24.72% in energy efficiency under different circumstances.

Virtual vehicle design has received a lot of attention recently and is commonly used to assist connected and autonomous vehicles (CAVs) in achieving cooperative merging. However, very few virtual vehicles have been designed to assist connected and human-driven vehicles (CHVs) in achieving cooperative control. This paper proposes an improved virtual vehicle design methodology that extends the application of the virtual vehicle concept to CHVs by considering driver compliance to ensure that CHVs and CAVs achieve cooperative merging. This method mainly includes two parts: the computation of vehicle cooperative state based on car-following model and the redesign of virtual vehicles. This paper also analyzes the stability of the mixed platoon composed of CHV and CAV, and obtains the conditions for the string stability of the mixed platoon. Simulation experiments demonstrate that the proposed method can overcome the problems caused by driver compliance in the collaborative process of the on-ramp area. Additionally, the proposed method has the advantages in reducing fuel consumptions and improving the comfort of drivers and passengers. The experiments based on SUMO platform show that the method can reduce traffic density and increase average speed, and that the improvement is more significant at lower CAV penetration rates. It will meet the real-time requirements of daily traffic by analyzing the computation time.

To minimize the stop-and-go behavior caused by human factors and traffic signal light constraints, this paper proposes a distributed MPC-based hierarchical cooperative control protocol for mixed vehicle groups consisting of Connected and Automated Vehicles (CAVs) and Human-driven Vehicles (HVs) in the vicinity of traffic signal lights (VTSL). Firstly, to characterize the formation pattern and mechanism of the mixed vehicle groups, the subgroup division and subgroup reorganization method is presented in the VTSL. Secondly, since human factors can affect the driving behavior of all vehicles within a subgroup, a HV model is established via considering human factors, such as the driver’s insensitivity to distance and speed of the preceding vehicle. Thirdly, to guarantee the maximum number of vehicles passing the VTSL and the minimum travel time under the traffic signal light constraints, a distributed MPC-based hierarchical cooperative control method is proposed from the Transportation Cyber-Physical System (T-CPS) perspective. Finally, the simulator experiment results indicate that the proposed control protocol is more advantageous and effective as the penetration rate of CAVs continuously increase.

Customized bus is considered an effective means to alleviate traffic congestion and reduce traffic-related environmental pollution caused by the increasing number of private cars. Exploring potential passenger information as the first stage of customized bus service has become a popular topic. Unlike manual investigation and passenger request methods, current studies utilize data mining methods to actively explore potential passengers from various historical travel data. However, the existing data mining methods only consider the spatiotemporal features of potential passengers and neglect the semantic features related to customized bus services, which play an important role in determining whether passengers are willing to use services. In this paper, we treated the exploration of potential customized bus passengers as a binary classification problem based on private car trajectory data. Then, we propose a novel data mining method, named iTrAdaboost-DTCN, which combines the strengths of deep learning and transfer learning. In detail, it integrates state-of-the-art deep neural networks by constructing a deep trajectory classification network (DTCN), which can automatically extract semantic feature representations to help improve classification accuracy. Due to the lack of city-wide labeled customized bus passenger information in practice, it also integrates instance-based transfer learning through improved TrAdaboost, which solves the learning problem of the target classification domain with limited labeled samples. Experimental results demonstrate that our method can explore potential passengers more effectively than other baseline methods. Furthermore, we apply our method to real-world scenarios and compare three travel characteristics of identified customized and non-customized bus passengers.

The reasonable and orderly traffic in vicinity of traffic signals is essential for the traffic efficiency of any signalized intersection. Under mixed traffic condition, adverse traffic impact of unreasonable vehicle merging behavior will be expected to be suppressed by designing active cyber-physical interaction mechanism in this section. From cyber-physical system (CPS) perspective, this paper proposes a vehicle group based merging control method for connected autonomous vehicles (CAVs). A vehicle group merging control protocol considering the physical driving disturbances is designed. Two kinds of stability proofs are given, which are asymptotic stability of the system with respect to vehicle internal disturbance and string stability of the system with respect to vehicle external disturbance. Simulation results show that through vehicle group merging control in vicinity of traffic signals, the proposed method can increase the traffic flow, average speed, and reduce traffic congestion in the vicinity of traffic signals.

Abstract Single target tracking based on computer vision helps to collect, analyse and exploit target information. The SwinTrack algorithm has received widespread attention as one of the twin network algorithms with the best trade‐off between tracking accuracy and speed, but it also suffers from the insufficient fusion of deep and shallow features leading to loss of shallow information and insufficient use of temporal information leading to inconsistency between target and template. Semantic information and detailed information are combined and multiple convolutional forms are introduced to propose a multi‐level feature fusion strategy to effectively fuse features in space. Besides, based on the idea of feedback, a dynamic template branching approach is also designed to fuse temporal features and enhance the representation of target features. The effectiveness of this method was verified on the OTB100 and GOT10K datasets.

Safely and efficiently completing unprotected left turns at intersections is challenging for both automated vehicles and human drivers, given that it is hard to predict the intentions of other road users. Currently, automated vehicles are inclined to adopt an overly conservative policy for safety reasons. And experienced drivers respond to right-of-way competition by employing “negotiation” skills to improve efficiency, mainly through steering, braking and acceleration. However, negotiations do not always go smoothly, and a phenomenon similar to “pedestrian face-off” is called “vehicle face-off”, specifically speaking, this host vehicle and vehicles involved in the competition perform the same maneuvers (acceleration or deceleration) continuously and simultaneously, leading to reduced efficiency and safety. In this paper, a new deep reinforcement learning (DRL) method is proposed based on deep convolutional fuzzy systems (DCFS) for automated vehicles to deal with unprotected left-turn scenarios on urban roads. A total of 30 subjects participated in the experiment, and the results show that the proposed method can provide human-like driving skills for automated vehicles, and effectively avoid “vehicle face-off” to improve the efficiency of unprotected left turns on the premise of ensuring safety.

In this paper, a novel lane keeping control method for automated vehicles based on human-simulated intelligent control (HSIC) is proposed, which is inspired by human expert drivers’ steering characteristics including good foresight, precise execution and notable intermittency. The novelty of the paper is to introduce the HSIC concept into lateral control of vehicles, which is a multi-mode control scheme implemented by the combination of the feedforward control for curve tracking and act-and-wait control for intermittent error correction. Theoretically, the stabilization problem of the HSIC method is investigated based on the switched system related method. Experiments on the joint simulation platform of PreScan and CarSim show that the newly presented HSIC scheme has better matching performance to the expert driver and good robustness. For automated lane keeping systems, the HSIC method could provide human-like qualities, which may be one of the essential points to determine whether the driver is comfortable or not when the driver hands over the steering authority, improve the transition smoothness in the scenario of human vehicle co-piloting, and eliminate the potential conflicts between manual driving and automated driving vehicles in the future mixed traffic flow.

The paper aims to ensure that traffic parameters are simultaneously available between sparse sensors, and to improve the traffic efficiency of expressways and achieve the consistent driving state of mixed vehicle groups. Firstly, we propose a dynamic mixed segmental linear (DMSL) traffic model with external disturbances including automated vehicles (AVs) and connected automated vehicles (CAVs) as the composition of mixed vehicle groups on the expressway. Secondly, an observer-based double closed-loop control strategy is investigated by jointly controlling the on-ramp flow in the outer-loop control system and the driving state of mixed vehicle groups on the main section of the expressway in the inner-loop control system. Thirdly, the convergence and stability of the inner-loop control system and the outer-loop control system are analyzed by using Lyapunov stability theory, and the control gain and observation gain of the double closed-loop are obtained via adopting linear matrix inequality (LMI) method. Finally, numerical simulation experiments are executed to confirm the feasibility and accuracy of the proposed control algorithm. The results illustrate that the proposed control method is effective while the percentage of automated vehicles (AVs) is gradually increasing.

To guarantee vehicle platoon driven pattern in heterogeneous nodes of mixed vehicle platoon (composed of connected and automated vehicles and human-driven vehicles, CAVs and HVs) on curved roads, this study develops a combined longitudinal and lateral controller, which comprises of selecting the key points (KPs) from the trajectory points of detected HVs, correcting the reference trajectory and controlling CAVs with the aid of the corrected KPs. To this end, a new concept, called KPs matrix, is proposed to manage the physical components of every KP by using image processing and vehicle-to-infrastructure (V2I) communication technology. Then, a trajectory correction scheme is presented to suppress the influence of nonstandard human-driven behavior by point set mapping approach in Real Variable Function theory. Furthermore, a novel controller is designed by incorporating the corrected KPs matrix and communication time delay. The stability and convergence of the proposed controller are rigorously analyzed based on the Lyapunov-Krasovskii stability theorem. In addition, extensive experiments are conducted to test the performance including three parts: the first part investigates the feasibility of the corrected KPs matrix by analyzing a video on high-way; the next part illustrates the control performance of the proposed controller on handling the cutting-corner issue (i.e. turning in advance), compared with the conventional controller. Meanwhile, the influence of time delay on the control performance is also analyzed in this study. The last implements driver-in-loop comparative experiments such that the performance of the proposed controller on eliminating the influence of nonstandard human-driven behavior is verified.

Abstract Vehicle detection plays an important role in intelligent transportation systems and security. Using the original Single Shot MultiBox Detector (SSD) directly for vehicle detection, lacks accuracy and stability. Moreover, most of the state‐of‐the‐art methods need cost a lot of time to inference. Vehicle detection is often used in complex traffic environments. Therefore, faster detection speed and higher detection accuracy are required. This study is aimed at developing a trade‐off between accuracy and speed vehicle detection framework based on the SSD algorithm. To improve the multi‐scale detection performance of SSD, semantic information, detailed features and receptive fields are combined to propose the feature pyramid enhancement strategy (FPES). On the other hand, the cascade detection mechanism is proposed to strengthen the positioning capability of SSD and an adaptive threshold acquisition method for object detection module (ODM) stage to improve model accuracy. Finally, a more efficient convolutional network is deployed through network slimming. Experimental results demonstrate that the proposed framework achieves state‐of‐the‐art performance on UA‐DETRAC and Udacity benchmarks. Interestingly, the inference time is the lowest for the proposed method than the state‐of‐the‐art methods, promising its application for fast and effective vehicle detection.

This article investigates the problem of multistability of almost-periodic solutions of Takagi-Sugeno fuzzy neural networks with nonmonotonic discontinuous activation functions and time-varying delays. Based on the geometrical properties of nonmonotonic activation functions, by using the Ascoli-Arzela theorem and the inequality techniques, it is demonstrated that under some reasonable conditions, the addressed networks have a locally exponentially stable almost-periodic solution in some hyperrectangular regions. We also estimate the attraction basins of the locally stable almost-periodic solutions, which indicates that the attraction basins of the locally exponentially stable almost-periodic solution can be larger than original hyperrectangular regions. These results, which include boundedness, globally attractivity, multiple stability, and attraction basins, generalize and improve the earlier publications, and can be extended to monostability and multistability of Takagi-Sugeno fuzzy neural networks with nonmonotonic discontinuous activation functions. Finally, several numerical examples are given to show the feasibility, the effectiveness, and the merits of the theoretical results.

This article is dedicated to designing a novel periodic impulsive control strategy for producing globally exponentially stable periodic solutions for switched neural networks with discrete and finite distributed time-varying delays. First, tunable parameters and cubic convex combination approach are proposed to study the globally exponential convergence of switched neural networks. Second, a sufficient criterion for the existence, uniqueness, and globally exponential stability of a periodic solution is demonstrated by using contraction mapping theorem and the impulse-delay-dependent Lyapunov-Krasovskii functional method. It is worth emphasizing that the addressed Lyapunov-Krasovskii functional covers both triple integral terms and novel quadruple integral terms, which makes the conservatism of the above criteria decrease. Even if the original neural network models are unstable or the impulsive effects are strong, the addressed neural network model can produce a globally exponentially stable periodic solution. These results here, which include boundedness, globally uniformly exponential convergence, and globally exponentially stability of the periodic solution, generalize and improve the earlier publications. Finally, two numerical examples and their computer simulations are given to show the effectiveness of theoretical results.

Since the unsaturating activation function is unbounded, more complex dynamics may exist in neural networks with this kind of activation function. In this article, monostability and multistability results of almost-periodic solutions are developed for fractional-order neural networks with unsaturating piecewise linear activation functions. Some globally Mittag-Leffler attractive sets are given, and the existence of globally Mittag-Leffler stable almost-periodic solution is demonstrated by using Ascoli-Arzela theorem. In particular, some sufficient conditions are provided to ascertain the multistability of almost-periodic solutions based on locally positively invariant set. It shows that there exists an almost-periodic solution in each positively invariant set, and all trajectories converge to this periodic trajectory in that rectangular area. Two illustrative examples are provided to demonstrate the effectiveness of the proposed sufficient criteria.

Recently feature pyramid composed of multi-level feature maps has been extensively used in region-free detectors to address multi-scale object detection. However, the contradiction between scale and context in the feature pyramid limits the detection performance, extraordinarily on small objects. Most works introduce an extra top-down path to overcome the limitation yet suffering from high computational burden. In this paper, we propose a novel Expansion Receptive Field Block (ERFB) to capture multiple strong contextual features at low computational cost, and then apply the Feature Attention Block (FAB) to eliminate the inconsistency between different features to generate more discriminative features. To be further, we construct an efficient and accurate detector (named CEBNet) mainly consists of Context Enhancement Blocks (CEBs), which are cascaded with ERFB and FAB. The extensive experiments on Pascal VOC and MS COCO demonstrate that CEBNet achieves state-of-the-art detection accuracy at a real-time processing speed.

Due to uneven illumination and dim environment in the tunnel, the monitored image is blurred, which makes it difficult to recognise the traffic status. Therefore, it is necessary to enhance the tunnel image in advance. In this study, a fast image enhancement algorithm based on imaging model constraint is proposed. First, the method uses the combination of global atmospheric light and partitioned atmospheric light to estimate the local atmospheric light. Second, the transmission is estimated based on the formula derived from the imaging model constraints. Third, the method uses a constant instead of illumination to balance tunnel image illumination. Last, the tunnel image is enhanced according to the imaging model. Experimental and comparative analysis results show that the proposed method can rapidly and effectively enhance the tunnel image.