With the development of artificial intelligence, deep neural network(DNN) tasks such as object detection have gradually integrated into production processes alongside traditional CPU tasks. Under certain conditions, these DNN tasks can be cooperatively inferred between terminal device (TD) and edge server (ES) to reduce the computational burden like traditional task offloading. However, existing task offloading models often define computational resources loosely for both edge servers and terminal devices. This leads to inaccurate task latency calculations due to scheduling delays from the operating system when the number of tasks exceeds available computing resources. To address these issues, we established a core-binding mechanism that considers task waiting delay when computational resources are fully occupied. We then built the Heterogeneous Task Offloading (HTO) model and measured computational latency and energy consumption in real-world scenarios. This optimization model is a mixed-integer programming problem that is difficult to solve. Therefore, we reformulated this problem as a Markov Decision Process (MDP). We propose a Heterogeneous Task Offloading Strategy based on the TD3 algorithm (HTO-TD3) to optimize both task delay and energy consumption. Numerical simulation results demonstrate that compared to existing reinforcement learning algorithms, the HTO-TD3 algorithm achieves superior convergence performance. Specifically, under certain load scenarios, the algorithm achieves a 38.3% reduction in latency and a 6.4% optimization in energy consumption simultaneously.

Integrated radar sensing and wireless power transfer (ISWPT) is an emerging paradigm that seeks to combine the functionalities of radar sensing and wireless power transfer into a unified system, utilizing a shared hardware platform to maximize resource efficiency. In this paper, we investigate the performance of an ISWPT system that is enhanced by extremely large-scale reconfigurable intelligent surfaces (XL-RIS), which serve to dynamically control the propagation environment and improve both radar sensing and wireless power transfer. Specifically, we consider a system where energy receivers are placed within the near-field region of the XL-RIS and investigate the joint optimization of beamforming at the base station and the XL-RIS reflection phases. The goal is to maximize the efficiency of wireless power transfer while simultaneously enhancing radar sensing performance. The formulated problem, though non-convex in nature, is efficiently addressed through the introduction of an alternating optimization algorithm. Numerical simulations demonstrate the effectiveness of the proposed algorithm, showing substantial improvements in both radar sensing accuracy and wireless power transfer performance compared to existing baseline schemes.

• A framework for cloud remanufacturing, encompassing a series of remanufacturing macroscopic phases, is established. • A multi-phase integrated scheduling problem for the proposed cloud remanufacturing system is introduced. • A mathematical model is developed to explain the scheduling problem. • An improved whale optimization algorithm integrating enhanced population updating mechanisms is designed to address this problem. The cloud remanufacturing system embraces a series of interdependent remanufacturing macroscopic phases (RMAs) with intricate precedence relationships, increasing the complexity of task scheduling and resource allocation. Thus, the multi-phase integrated scheduling is necessary to manage remanufacturing tasks and optimize resources and capabilities effectively in the cloud environment. This research investigates the multi-phase integrated scheduling problem for cloud remanufacturing system involving a series of RMAs including initial inspection, disassembly, reprocessing, reassembly, and final test. A mathematical model is created to explain the scheduling issue using the suggested cloud remanufacturing framework. Due to the high complexity of integrated scheduling, traditional meta -heuristic algorithms cannot be directly applied to solving the problem. Thus, an improved whale optimization algorithm (IWOA) incorporating the self-adaptive weighting and quadratic interpolation techniques is proposed for addressing the studied problem efficiently. A case study is designed and conducted, and the findings indicate that the IWOA is more effective than other methods in addressing the proposed complex scheduling issues with better accuracy, faster computation, and improved convergence efficiency.

Large-scale wires and cables (W&Cs) are the nerves of large and complex electromechanical products vital to their regular operation. The laying process of W&C has been highly dependent on manual work and urgently needs intelligent guidance like augmented reality (AR). However, model registration and occlusion handling based on AR for large-scale W&C laying scenes cannot achieve high-quality results to date because of the lack of texture, the local field of vision, and other characteristics. Therefore, a markerless AR guidance method for large-scale W&C laying is proposed to address this long-lasting problem. First, camera, ultrawideband, and inertial measurement unit sensors are integrated to coarsely register the W&C models based on coordinate transformation and mapping virtual and physical spaces. Then, a local edge matching method is used to register the W&C models based on the coarse one finely. Next, the occlusion relationships of the W&C models are handled based on assembly constraints; the models are anchored in the simultaneous localization and mapping map for visual and continuous guidance. A locomotive W&C laying case study and evaluation results show that this method meets industrial application requirements regarding efficiency, accuracy, and robustness and has several advantages compared to other methods, including extra-large scale, automatic registration, and virtual-real fusion.

Abstract Depression diagnosis based on speech signals has the advantages of non‐invasiveness, low cost, and few restrictions on portability. The research on the recognition of the depression state is carried out based on the acoustic information in the speech signal. Aiming at the interview dialogue speech in the consultation environment, a hierarchical attention temporal convolutional network (HATCN) acoustic depression recognition model is proposed. For sentence acoustic feature learning, a regional attention mechanism is introduced to extract multi‐scale sentence features; for segment acoustic feature extraction, the traditional attention mechanism is used to calculate, which is in line with human cognitive mechanism. In addition, a periodic focal loss function is introduced to address the imbalance of positive and negative samples in depression diagnosis. Experiments show that the proposed acoustic depression recognition model has a certain improvement in recognition performance compared with other methods. At the same time, the influence of noise on the recognition of acoustic depression in the real consultation environment is analysed through experiments, and the data enhancement is carried out utilising speech noise, which proves the effectiveness of the data expansion of speech noise.

This article is devoted to the fuzzy fault-tolerant tracking control of Markov jump systems with unknown mismatched faults. To reconstruct the faults and system states, a sequence of proportional–integral observers are established via the system outputs. With the help of a structure separation technique, the proportional–integral gains and the observer gains are solved by a unified linear matrix inequality framework. Resorting to the rebuilt faults and states from an iterative estimation algorithm, a backstepping-based fuzzy fault-tolerant tracking control scheme against the mismatched faults is established to make the resultant closed-loop system be uniformly ultimately bounded. Simulations are provided to verify the effectiveness of the proposed methods.

This paper proposes a new type of product service system (PSS) design method named Service-oriented Design (SOD), the design method aim at building a service-based modeling system to totally satisfy customer demand. The mainline of SOD is customer demand acquire, service function and performance, service structure and service activity. Based on this method and its mainline, we propose a CAD system (Service Design System) that designer is able to design a service using this visible system. Through its application to business case such as hydraulic fracturing, the Service Design System was proven to support designer to design a steady and efficient service.