The rapid development of advanced sensing and artificial intelligence technologies, has advanced autonomous driving (AD) systems by providing intelligent route planning decisions. However, how to construct an interpretable and efficient decision-making method that can adapt to various complex driving scenarios has become an important and challenging research topic. In this article, a knowledge graph (KG) for AD systems is constructed based on heterogeneous data such as traffic rules and network information. A deep learning model combining bidirectional long short-term memory and conditional random field is used to achieve joint learning of entity recognition and relationship extraction. In order to make decisions on driving behaviors, this article introduces a deep reinforcement learning framework designed to perform knowledge reasoning over the driving KG, which integrates an integrated reward function and an action dropout mechanism. Experimental comparisons against the other advanced knowledge reasoning algorithms on a practical driving rules dataset validate the effectiveness and advantages of the proposed method, with an overall mean average precision exceeding 94%. The validity of the proposed method has also been verified on the simulation platform in different road scenarios such as multilane, roundabout, intersection and thru-junction.

When autonomous vehicles (AVs) encounter sudden hazards, how to take the correct collision-free measures in emergencies has always been a focus of research. The reachable sets method has been widely applied in local path planning methods for AVs to address the collision-free path planning problem for dense, dynamic obstacles in traffic scenarios. However, the reachable sets method is limited by its inherent tendency to overestimate obstacles, making it difficult to quantify the probability of conflicts between other traffic participants and AVs. Furthermore, path planning within the reachable set may encounter local minima, potentially resulting in planning failure.This paper proposes a reachable probabilistic distribution (RPD) method, integrating reachable sets with the Interactive Multiple Model Filtering (IMM) method. The proposed method begins by collecting surrounding environmental data using Vehicle-to-Everything (V2X) technology. It then models both the obstacles and the ego vehicle, followed by constructing the occupancy sets of the obstacles and the reachable set of the ego vehicle. IMM then predicts obstacle behavior to determine their probabilities of being in each lane at future intervals. These probabilities are combined with obstacles’ occupancy sets and ego vehicle’s reachable sets to derive RPD. Compared to traditional reachable sets, the RPD framework leverages probabilistic information to simplify candidate path evaluation, greatly enhancing computational efficiency. To avoid local minima failures in local path planning within RPD, a method combining discrete strategies with Model Predictive Control (MPC) is proposed. The proposed method offers a safety reference for intelligent transportation systems based on V2X technology. It employs discrete strategies to achieve comprehensive coverage of feasible local paths, while MPC is used for real-time tracking. Validated in AVs, this method ensures robust collision avoidance with stability and safety, demonstrating its applicability in intelligent transportation systems.

A systematic investigation of evolving visual cryptography scheme (EVCS) is carried out in this article. The evolving scheme, denoted as \((k,\infty)\) , differs from the \((k,n)\) threshold in that it permits an arbitrary and perhaps unlimited number of participants. More importantly, the access structure can be updated dynamically by adding new users. First of all, a preliminary implementation strategy for the \((2,\infty)\) EVCS is introduced. Then, by employing the \((2,2)\) VCS recursively with the \((2,\infty)\) EVCS, a \((k,\infty)\) EVCS is created. In order to enhance the performance, an improved scheme is constructed based on the multi-secret VCS (MVCS) and a series of EVCS schemes with thresholds of \((1,\infty)\) , \(\cdots\) , \((k-1,\infty)\) . Moreover, Boolean XOR operation is adopted for secret recovery to further improve the visual quality. To facilitate the XOR decryption, a novel access structure partition algorithm is presented. Additionally, the proposed partition method can successfully solve the security issue in existing multi-secret XOR-based VCS (MXVCS). By integrating the more secure MXVCS into the improved scheme, XOR decryption is provided. The two proposed methods are shown to be effective and advantageous through extensive experiments and comparisons.

Safety is a paramount concern in the realm of autonomous vehicles. Developing precise safety assessment is challenging due to the need to blend qualitative and quantitative analyses of various safety factors. To address this challenge, this paper presents an innovative safety assessment method based on the cloud model. This method employs fundamental cloud model elements like expectation, entropy, and ultra-entropy. It also employs a sophisticated double conditional single rule generator to integrate multiple assessment indicators, resulting in an integrated risk assessment cloud. This cloud dynamically represents varying risk levels based on indicator characteristics. The method evaluates the real-time safety level by assessing the proximity between the integrated risk assessment cloud and the standard cloud. This proximity analysis reveals the prevailing risk level. Empirical validation involves rigorous testing within typical scenarios, demonstrating the utility and potential of the method to assess safety for decision-making of autonomous vehicles. The capacity of the method to monitor and assess autonomous vehicle decision-making systems makes it a significant contribution to the field. Beyond empirical contributions, this paper offers theoretical insights that can shape the future of safety assessment methods for autonomous vehicles. In summary, this paper emphasizes the importance of safety for autonomous vehicles and paves the way for evolving safety assessment methods in this dynamic field.

In this research, we consider the evolving threshold access structure, denoted as$(k, \infty)$, for size invariant visual cryptography scheme (SIVCS). The so-called$(k, \infty)$threshold indicates the number of participants is supposed to be infinite and the access structure would be dynamically adjusted at any time by adding or deleting participants. First of all, the concept and definition of$(k, \infty)$-SIVCS are described. Shadow construction, constituted by random number generators and their choosing probabilities, for the$(k, \infty)$-SIVCS is then given. A contrast-maximizing problem for determining the generators and choosing probabilities is built based on the$(k, \infty)$-SIVCS. A simulated annealing-based algorithm is introduced to solve the optimization problem. The best solution from the simulated annealing-based algorithm forms a feasible$(k, \infty)$-SIVCS. To further improve the visual quality, a$(k, \infty)$-SIVCS using Boolean XOR decryption is also presented. Experimental results and comparisons are shown, demonstrating that the proposed techniques are feasible and advanced in the aspects of shadow size and contrast.

Image matting is a fundamental technique in visual understanding and has become one of the most significant capabilities in mobile phones. Despite the development of mobile storage and computing power, achieving diverse mobile Artificial Intelligence Generated Content (AIGC) applications remains a great challenge. To address this issue, we present an innovative demonstration of an automatic system called "Matting Moments" that enables automatic image editing based on matting models in different scenarios. Coupled with accurate and refined matting subjects, our system provides visual element editing abilities and backend services for distribution and recommendation that respond to emotional expressions. Our system comprises three components: 1) photo content structuring, 2) data-driven matting engine, and 3) AIGC functions for generation, which automatically achieve diverse photo beautification in the gallery. This system offers a unified framework that guides consumers to obtain intelligent recommendations with beautifully generated contents, helping them enjoy the moments and memories of their present life.

Nowadays, human-oriented has already become the direction of the development of the intelligent vehicle, among which, the cabin, in constant contact with drivers, is getting more and more attention. Intelligent assisted systems have alleviated the burden on drivers during long journeys and provided a remedy for operational errors. As the trend towards increasingly intelligent vehicles, the issue of human-machine co-driving is receiving attention from scientific researchers. The technologies of human-machine interactions usually contain two parts, the human-to-vehicle and vehicle-to-human. This paper analyzes the potential innovation of human-machine systems from the perspective of human sensing, including visual, auditory, tactile, and olfactory. Based on the review of human-machine technologies, the current intelligentization of vehicles is divided into driver interaction and crew service systems. Then, the structure of a future intelligent interaction system considering multi-sensing is proposed and further discussed. Finally, by analyzing the relationship between the system for human and autonomous systems, a classification of the intelligence level for interaction systems is presented.

5G networks need to support various use cases which may require different quality of service (QoS). Network slicing (NS) is an innovative concept to customize different logical networks from a common general infrastructure for these services. A slice is composed of a set of virtual network functions (VNFs) that can be deployed on standard commodity servers as virtual machines (VMs). However, the packet delay of VMs may be affected by other VMs on the same server. In this paper, we investigate network slice deployment policy to improve the network performance by guaranteeing the latency requirements. Moreover, a realistic mathematical model is formulated, considering the effects of virtualization on end-to-end delay. Due to the diversity of network slices' topologies and various performance requirements, we propose a heuristic algorithm to deploy different network slices. Extensive experiments demonstrate that our algorithm can get solutions with reasonable execution time and achieve lower delay and higher acceptance ratio.