As urban traffic condition is diverse and complicated, applying reinforcement learning to reduce traffic congestion becomes one of the hot and promising topics. Especially, how to coordinate the traffic light controllers of multiple intersections is a key challenge for multi-agent reinforcement learning (MARL). Most existing MARL studies are based on traditional Q-learning, but unstable environment leads to poor learning in the complicated and dynamic traffic scenarios. In this paper, we propose a novel multi-agent recurrent deep deterministic policy gradient (MARDDPG) algorithm based on Deterministic Policy Gradient algorithm for traffic light control (TLC) in vehiclar networks. Specifically, the centralized learning in each critic network enables each agent to estimate the policies of other agents in the decision-making process and each agent can coordinate with each other, alleviating the problem of poor learning performance caused by environmental instability. The decentralized execution enables each agent to make decisions independently. We share parameter in actor networks to speed up training process and reduce the memory footprint. The addition of LSTM is beneficial to alleviate the instability of the environment caused by partial observable state. We utilize surveillance cameras and vehicular networks to collect status information for each intersection. Unlike previous work, we have not only considered the vehicle but also considered the pedestrians waiting to pass through the intersection. Moreover, we also set different priorities for buses and ordinary vehicles. The experimental results in a vehicular network show that our method can run stably in various scenarios and coordinate multiple intersections, which significantly reduces vehicle congestion and pedestrian congestion.

This paper presents the first study on scheduling for cooperative data dissemination in a hybrid infrastructure-to-vehicle (I2V) and vehicle-to-vehicle (V2V) communication environment. We formulate the novel problem of cooperative data scheduling (CDS). Each vehicle informs the road-side unit (RSU) the list of its current neighboring vehicles and the identifiers of the retrieved and newly requested data. The RSU then selects sender and receiver vehicles and corresponding data for V2V communication, while it simultaneously broadcasts a data item to vehicles that are instructed to tune into the I2V channel. The goal is to maximize the number of vehicles that retrieve their requested data. We prove that CDS is NP-hard by constructing a polynomial-time reduction from the Maximum Weighted Independent Set (MWIS) problem. Scheduling decisions are made by transforming CDS to MWIS and using a greedy method to approximately solve MWIS. We build a simulation model based on realistic traffic and communication characteristics and demonstrate the superiority and scalability of the proposed solution. The proposed model and solution, which are based on the centralized scheduler at the RSU, represent the first known vehicular ad hoc network (VANET) implementation of software defined network (SDN) concept.

A multifactorial evolutionary algorithm (MFEA) is a recently proposed algorithm for evolutionary multitasking, which optimizes multiple optimization tasks simultaneously. With the design of knowledge transfer among different tasks, MFEA has demonstrated the capability to outperform its single-task counterpart in terms of both convergence speed and solution quality. In MFEA, the knowledge transfer across tasks is realized via the crossover between solutions that possess different skill factors. This crossover is thus essential to the performance of MFEA. However, we note that the present MFEA and most of its existing variants only employ a single crossover for knowledge transfer, and fix it throughout the evolutionary search process. As different crossover operators have a unique bias in generating offspring, the appropriate configuration of crossover for knowledge transfer in MFEA is necessary toward robust search performance, for solving different problems. Nevertheless, to the best of our knowledge, there is no effort being conducted on the adaptive configuration of crossovers in MFEA for knowledge transfer, and this article thus presents an attempt to fill this gap. In particular, here, we first investigate how different types of crossover affect the knowledge transfer in MFEA on both single-objective (SO) and multiobjective (MO) continuous optimization problems. Furthermore, toward robust and efficient multitask optimization performance, we propose a new MFEA with adaptive knowledge transfer (MFEA-AKT), in which the crossover operator employed for knowledge transfer is self-adapted based on the information collected along the evolutionary search process. To verify the effectiveness of the proposed method, comprehensive empirical studies on both SO and MO multitask benchmarks have been conducted. The experimental results show that the proposed MFEA-AKT is able to identify the appropriate knowledge transfer crossover for different optimization problems and even at different optimization stages along the search, which thus leads to superior or competitive performances when compared to the MFEAs with fixed knowledge transfer crossover operators.

The Internet of Vehicles (IoV) has become a fundamental platform for advancing Intelligent Transportation Systems (ITSs) and Intelligent Connected Vehicles (ICVs). However, the increasing volume of data generated by vehicle sensors and the computational demands of Artificial Intelligence (AI) algorithms present significant challenges for the platform. Edge Intelligence (EI), which brings intelligent computing and data processing closer to vehicles, has emerged as a potential solution. In this survey, we provide a comprehensive overview of Edge Intelligence for the Internet of Vehicles. We begin by discussing the motivations behind employing EI in the IoV for typical AI computations. To fully exploit the potential of EI in heterogeneous IoV environments, we present a layered vehicular EI architecture and discuss its benefits and challenges. Furthermore, we provide a taxonomy of EI approaches for vehicular networks, focusing on cooperative inference, distributed training, and collaborative sensing, in terms of their schemas and advanced frameworks. Finally, we explore emerging trends and research directions in this field, including vehicle-road-cloud integration, generative AI-driven IoV, and vehicular cyber-physical fusion. By offering insights into state-of-the-art techniques and trends, this survey aims to enable researchers to develop innovative solutions for transforming the intelligent IoV ecosystem.

Accurate positioning is a key factor for enabling innovative applications in intelligent transportation systems. Cutting-edge communication technologies make cooperative localization a promising approach for accurate vehicle positioning. In this paper, we first propose a ranging technique called weighted least squares double difference (WLS-DD), which is used to detect intervehicle distances based on the sharing of GPS pseudorange measurements and a weighted least squares method. It takes the carrier-to-noise ratio (CNR) of raw pseudorange measurements into consideration for mitigating noises so that it can improve the accuracy of the distance detection. We show the superiority of WLS-DD by conducting a series of field experiments. Based on intervehicle distance detection, we propose a distributed location estimate algorithm (DLEA) to improve the accuracy of vehicle positioning. The implementation of DLEA only relies on inaccurate GPS pseudorange measurements and the obtained intervehicle distances without using any reference points for positioning correction. Moreover, to evaluate the joint effect of WLS-DD and DLEA, we derive a data fitting model based on the observed distance detection bias from field experiments, which generates parameters in a variety of environments for performance evaluation. Finally, we demonstrate the effectiveness of the proposed solutions via a comprehensive simulation study.

The vast demand for diverse applications raises new networking challenges, which have encouraged the development of a new paradigm of Internet of Things (IoT), e.g., LoRa. LoRa is a proprietary spread spectrum modulation technique that provides a solution for long-range and ultra-low power-consumption transmission. Due to promising prospects of LoRa, significant effort has been made on this compelling technology since its emergence. In this article, we provide a comprehensive survey of LoRa from a systematic perspective: LoRa analysis, communication, security, and its enabled applications. First, we summarize works focusing on analyzing the performance of LoRa networks. Then, we review studies enhancing the performance of LoRa networks in communication. Afterward, we analyze the security vulnerabilities and countermeasures. Finally, we survey the various LoRa-enabled applications. We also present comparisons of existing methods, together with insightful observations and inspiring future research directions.