Advanced driver assistance systems (ADASs) enhance driving safety and convenience by providing auxiliary functions. However, traditional rule-based or learning-based ADAS lack the capability for commonsense-based environmental understanding and multisensor data fusion, which leads to limitations in complex dynamic environments. Multimodal large language models (MLLMs) can effectively integrate data from different modalities and possess strong environmental perception and commonsense reasoning abilities, offering more intelligent driver assistance services within Internet of Things (IoT) networks. In this article, we propose a cloud-edge collaborative ADAS based on MLLMs, utilizing IoT networks by deploying a smaller model, CogVLM2, at the edge and a larger model, ChatGPT-4o, in the cloud to achieve collaborative driver assistance services. Specifically, we first reannotate the BDD-X dataset and use it to fine-tune CogVLM2 with LoRA, while applying few-shot learning to ChatGPT-4o to enhance their understanding and decision-making capabilities in traffic scenarios. We then formulate service latency, energy consumption, and Quality-of-Service (QoS) models for the cloud-edge collaborative ADAS in IoT networks, optimizing the combination of these models. Finally, we design an improved DDPG-based task offloading algorithm by introducing a multistep reward mechanism and using a diffusion model to generate noise, aiming to determine the optimal execution location (i.e., cloud, edge, or local) for each task. Experimental results show that both CogVLM2 and ChatGPT-4o can achieve basic ADAS functionality. After fine-tuning and few-shot learning, their task success rates were significantly improved. Moreover, compared to other mainstream deep reinforcement learning-based task offloading algorithms, the improved DDPG task offloading algorithm demonstrates better performance in latency, energy consumption, and QoS within IoT networks.

Embodied Artificial Intelligence (AI) bridges the cyberspace and the physical space, driving advancements in autonomous systems like theVehicularEmbodiedAINETwork (VEANET). VEANET integrates advanced AI capabilities into vehicular systems to enhance autonomous operations and decision-making. Embodied agents, such as Autonomous Vehicles (AVs), are autonomous entities that can perceive their environment and take actions to achieve specific goals, actively interacting with the physical world. Embodied Agent Twins (EATs) are digital models of these embodied agents, with various Embodied Agent AI Twins (EAATs) for intelligent applications in cyberspace. In VEANETs, EAATs act as in-vehicle AI assistants to perform diverse tasks supporting autonomous driving using generative AI models. Due to limited onboard computational resources, AVs offload EAATs to nearby RoadSide Units (RSUs). However, the mobility of AVs and limited RSU coverage necessitates dynamic migrations of EAATs, posing challenges in selecting suitable RSUs under information asymmetry. To address this, we construct a multi-dimensional contract theoretical model between AVs and alternative RSUs. Considering that AVs may exhibit irrational behavior, we utilize prospect theory instead of expected utility theory to model the actual utilities of AVs. Finally, we employ a Generative Diffusion Model (GDM)-based algorithm to identify the optimal contract designs, thus enhancing the efficiency of EAAT migrations. Numerical results demonstrate the superior efficiency of the proposed GDM-based scheme in facilitating EAAT migrations compared with traditional deep reinforcement learning methods.

The synergy between Unmanned Aerial Vehicles (UAVs) and metaverses is giving rise to an emerging paradigm named UAV metaverses, which create a unified ecosystem that blends physical and virtual spaces, transforming drone interaction and virtual exploration. UAV Twins (UTs), as the digital twins of UAVs that revolutionize UAV applications by making them more immersive, realistic, and informative, are deployed and updated on ground base stations, e.g., RoadSide Units (RSUs), to offer metaverse services for UAV Metaverse Users (UMUs). Due to the dynamic mobility of UAVs and limited communication coverages of RSUs, it is essential to perform real-time UT migration to ensure seamless immersive experiences for UMUs. However, selecting appropriate RSUs and optimizing the required bandwidth is challenging for achieving reliable and efficient UT migration. To address the challenges, we propose a tiny machine learning-based Stackelberg game framework based on pruning techniques for efficient UT migration in UAV metaverses. Specifically, we formulate a multi-leader multifollower Stackelberg model considering a new immersion metric of UMUs in the utilities of UAVs. Then, we design a Tiny Multi-Agent Deep Reinforcement Learning (Tiny MADRL) algorithm to obtain the tiny networks representing the optimal game solution. Specifically, the actor-critic network leverages the pruning techniques to reduce the number of network parameters and achieve model size and computation reduction, allowing for efficient implementation of Tiny MADRL. Numerical results demonstrate that our proposed schemes have better performance than traditional schemes.

Vehicular Metaverses represent emerging paradigms arising from the convergence of vehicle road cooperation, Metaverse, and augmented intelligence of things. Users engaging with Vehicular Metaverses (VMUs) gain entry by consistently updating their Vehicular Twins (VTs), which are deployed on RoadSide Units (RSUs) in proximity. The constrained RSU coverage and the consistently moving vehicles necessitate the continuous migration of VTs between RSUs through vehicle road cooperation, ensuring uninterrupted immersion services for VMUs. Nevertheless, the VT migration process faces challenges in obtaining adequate bandwidth resources from RSUs for timely migration, posing a resource trading problem among RSUs. In this paper, we tackle this challenge by formulating a game-theoretic incentive mechanism with multi-leader multi-follower, incorporating insights from social-awareness and queueing theory to optimize VT migration. To validate the existence and uniqueness of the Stackelberg Equilibrium, we apply the backward induction method. Theoretical solutions for this equilibrium are then obtained through the Alternating Direction Method of Multipliers (ADMM) algorithm. Moreover, owing to incomplete information caused by the requirements for privacy protection, we proposed a multi-agent deep reinforcement learning algorithm named MALPPO. MALPPO facilitates learning the Stackelberg Equilibrium without requiring private information from others, relying solely on past experiences. Comprehensive experimental results demonstrate that our MALPPO-based incentive mechanism outperforms baseline approaches significantly, showcasing rapid convergence and achieving the highest reward.

To facilitate the implementation of deep learning-based vehicular applications, vehicular federated learning is introduced by integrating vehicular edge computing with the newly emerged federated learning technology. In vehicular federated learning, it is widely considered that the raw data collected by vehicles have complete ground-truth labels. This, however, is not realistic and inconsistent with the current applications. To deal with the above dilemma, a semisupervised vehicular federated learning (Semi-VFL) framework is proposed. In the framework, each vehicular client uses labeled data shared by an application provider, and its own unlabeled data to cooperatively update a global deep neural network model. Furthermore, the application provider combines the multidimensional contract theory with prospect theory (PT) to design an incentive mechanism to stimulate appropriate vehicular clients to participate in Semi-VFL. Multidimensional contract theory is used to deal with the information asymmetry scenario where the application provider is not aware of vehicular clients’ 3-D cost information, while PT is used to model the application provider’s risk-aware behavior and make the incentive mechanism more acceptable in practice. After that, a closed-form solution for the optimal contract items under PT is derived. We present the real-world experimental results to demonstrate that Semi-VFL achieves the advantages in both the test accuracy and convergence speed, in comparison with existing baseline schemes. Based on the experimental results, we further perform the simulations to verify that our incentive mechanism is efficient.

The 5th generation (5G) mobile network technology is accelerating the development of autonomous vehicles by significantly shortening the communication latency and improving the reliability of network connection and transmission. However, as the number of vehicles increases, neither cloud servers nor multi-access edge computing (MEC) servers alone could sufficiently meet the Quality-of-Service (QoS) requirements for computing-intensive vehicle tasks. In this paper, we consider a hierarchical offloading scenario, where vehicle tasks are allowed to execute in MEC servers, convergence servers or cloud servers. To reduce the cost of latency and energy, we optimize the communication and computation resource allocation problem. The optimization problem is converted to a Markov decision process, and deep reinforcement learning is used to tackle the resource slicing and scheduling problem. Simulation results show that the proposed scheme is more resilient and efficient than that of single cloud server offloading or single MEC server offloading.

Federated learning is a promising tool in the Internet-of-Things (IoT) domain for training a machine learning model in a decentralized manner. Specifically, the data owners (e.g., IoT device consumers) keep their raw data and only share their local computation results to train the global model of the model owner (e.g., an IoT service provider). When executing the federated learning task, the data owners contribute their computation and communication resources. In this situation, the data owners have to face privacy issues where attackers may infer data property or recover the raw data based on the shared information. Considering these disadvantages, the data owners will be reluctant to use their data to participate in federated learning without a well-designed incentive mechanism. In this article, we deliberately design an incentive mechanism jointly considering the task expenditure and privacy issue of federated learning. Based on a differentially private federated learning (DPFL) framework that can prevent the privacy leakage of the data owners, we model the contribution as well as the computation, communication, and privacy costs of each data owner. The three types of costs are data owners' private information unknown to the model owner, which thus forms an information asymmetry. To maximize the utility of the model owner under such information asymmetry, we leverage a 3-D contract approach to design the incentive mechanism. The simulation results validate the effectiveness of the proposed incentive mechanism with the DPFL framework compared to other baseline mechanisms.

In industrial Internet of things (IIoT), peer-to-peer (P2P) energy trading ubiquitously takes place in various scenarios, e.g., microgrids, energy harvesting networks, and vehicle-to-grid networks. However, there are common security and privacy challenges caused by untrusted and nontransparent energy markets in these scenarios. To address the security challenges, we exploit the consortium blockchain technology to propose a secure energy trading system named energy blockchain. This energy blockchain can be widely used in general scenarios of P2P energy trading getting rid of a trusted intermediary. Besides, to reduce the transaction limitation resulted from transaction confirmation delays on the energy blockchain, we propose a credit-based payment scheme to support fast and frequent energy trading. An optimal pricing strategy using Stackelberg game for credit-based loans is also proposed. Security analysis and numerical results based on a real dataset illustrate that the proposed energy blockchain and credit-based payment scheme are secure and efficient in IIoT.

With the rapid increase of mobile devices, the computing load of roadside cloudlets is fast growing. When the computation tasks of the roadside cloudlet reach the limit, the overload may generate heat radiation problem and unacceptable delay to mobile users. In this paper, we leverage the characteristics of buses and propose a scalable fog computing paradigm with servicing offloading in bus networks. The bus fog servers not only provide fog computing services for the mobile users on bus, but also are motivated to accomplish the computation tasks offloaded by roadside cloudlets. By this way, the computing capability of roadside cloudlets is significantly extended. We consider an allocation strategy using genetic algorithm (GA). With this strategy, the roadside cloudlets spend the least cost to offload their computation tasks. Meanwhile, the user experience of mobile users are maintained. The simulations validate the advantage of the propose scheme.

The limited lifetime of wireless nodes has become the essential bottleneck of system performance and wide-scale deployment of wireless networks. In this paper, we consider a practical mobile chargeable network in which each single mobile charger is able to charge multiple target nodes simultaneously. To tackle the problem, the cooperative grouping of the wireless nodes is proposed to reduce the number of traversing spots of the mobile chargers. Meanwhile, the cooperative charging among the mobile chargers is studied to conserve their energy consumption. The numerical results show that the proposed scheme outperforms existing strategies in both even-density and uneven-density wireless networks.

In recent years, an intelligent noninvasive endoscope has been felt necessary for early diagnosis of malignant tumor in gastrointestine (GI). It is our purpose to develop a microwireless robotic endoscope to examine the human GI. The presented robot's diameter and length is 10 and 190 mm, respectively. A locomotion principle based on biomimetic earthworm is adopted for a higher adaptability to the GI. The robot is composed of three linear driving cells. A micromotor, a reducer, and a microscrew pair mechanism are integrated into each cell. Multijoints are used to connect every two driving cells for robot's high flexibility. The robot's energy is continuously supplied in real time by an energy transmitting system based on electromagnetic coupling. Experiments on the energy transmission and locomotion are performed to test the power transferring volume and locomotion effect in GI. The experiments indicate that the minimum received power, 400 mW, is obtained in a cylindrical space with 200 mm diameter and 200 mm length. In vitro experiments in a pig intestine indicate that this robot can move forward or backward effectively. This paper provides a good prototype for the deeper research on locomotion theory and energy transferring technology in vivo in future.