With the rapid evolution of the Industrial Internet of Things (IIoT), industrial networks are required to support massive industrial devices with bounded low-latency transmission. To address these requirements, integrating the fifth-generation (5G) with time-sensitive networking (TSN) has been proposed. However, existing joint resource allocation methods struggle to achieve seamless low-latency deterministic scheduling between the 5G system (5GS) and TSN networks. Focusing on large-scale uplink transmission scenarios, a base station (gNB)-driven joint time-frequency resource allocation architecture is proposed. Within this architecture, the 5GS is modeled as a TSN bridge seamlessly integrated with the TSN cyclic queuing and forwarding (CQF) mechanism. Existing joint resource allocation algorithms suffer from local optima, low computational efficiency, and inability to capture global time-triggered (TT) flow interactions for globally optimal solutions. Accordingly, a gNB-driven multi-agent proximal policy optimization (MAPPO)-based joint time-frequency resource allocation algorithm, termed gNB-DMJRA, is further proposed. This algorithm adopts the centralized training and distributed execution (CTDE) framework, leveraging global information to better coordinate the scheduling of multiple TT flows and avoid convergence to local optimal solutions. In addition, the periodicity of TT flows is exploited to reduce computational complexity and the action space, thereby further improving learning efficiency and convergence. Simulation results demonstrate that under 1000 TT flows, the proposed algorithm reduces maximum latency by up to 74.27%, improves the scheduling success rate by 82.15%-331.75%, and achieves faster convergence than benchmarks, confirming its effectiveness and efficiency for large-scale TT flow scheduling.

The computing power network (CPN) serves as a key infrastructure for future networks, facilitating the connection of ubiquitous computing resources distributed across various locations. The continuous emergence of computation-intensive and delay-sensitive applications highlights the crucial need to fully utilize limited computing resources and the importance of building a resilient communication network. Primary-backup (PB) based transmission is a commonly used technique to enhance network reliability. However, implementing this approach in CPN with a consideration of load balancing introduces significant complexity and has received limited research attention. In this paper, we designed a deterministic computing power network (Det-CPN) architecture based on segment routing over IPv6 (SRv6). On top of the above architecture, we proposed a best computing node selection method based on a comprehensive index calculation and ranking (CICR) algorithm to determine the optimal computing node for task transmission. Subsequently, we developed a bandwidth sharing-based multi-path transmission (BSMT) algorithm to realize the maximization of the system efficiency. Simulation results demonstrate that in adverse network conditions (overloaded with a failure rate of 0.02), compared to the traditional dual-path redundant forwarding mechanism, the proposed solution achieves an average reduction of 22.3% in transmission latency, an average improvement of 39.94% in task success rate, a decrease of 19.4% in bandwidth occupation rate, and an increase of 31.05% in computing resource utilization rate.

Time Sensitive Networking (TSN) has been widely considered as a promising networking technology in industrial fields as its capibility of deterministic transmission. One of the main challenges in TSN application is the complexity of network planning, including topology design, flow routing and scheduling schemes. In these three tasks, topology design plays a vital role in reducing costs and supporting the feasibility of routing and scheduling schemes. While recent researchers make some progress in routing and scheduling algorithms, most studies lack effective approaches for optimizing network topology, limiting the practical applicability of TSN. This paper addresses this problem by presenting a joint design method (JDM) for low-cost TSN topology design while also ensuring the feasibility of flow routing and scheduling. A unified mathematical model is developed to integrate TSN topology, routing and scheduling into a joint optimization problem, minimizing the overall network cost. On this basis, a one-hot vectorization technique is applied to linearize scheduling constraints to enhance the computational efficiency. Simulation results show that compared with other methods, the proposed JDM generates TSN topology at the lowest cost, ensuring flows deterministic transmission within minutes.

To address the limitations of existing Time-Aware Shaper (TAS) scheduling strategies in terms of configuration complexity and resource utilization, this paper proposes a Density-Adaptive Gate Control (DAGC) method, designed to meet the strict latency requirements of high-priority traffic while improving bandwidth utilization. We develop a system model for industrial network environments and design a corresponding scheduling model. By analyzing key performance indicators such as end-to-end latency, jitter, and packet loss rate, we demonstrate the effectiveness of the proposed method. Then, we further demonstrate through MATLAB simulations and real-network experiments that the DAGC method significantly outperforms two existing scheduling methods. In scenarios where traditional end devices cannot alter task offset times, DAGC reduces latency by 98.81% and the number of guard bands by 25.65%, while ensuring zero packet loss. Under network congestion, DAGC decreases the latency of bursty emergency traffic by 89.18%, while ensuring that high-priority traffic remains unaffected by low-priority traffic and maintaining zero packet loss even under heavy load.

Industrial time-sensitive networking (TSN) is pivotal for ensuring real-time and reliable flow transmission. There is a growing focus on its scalable scheduling for time-critical flows pursuing ultralow latency and jitter. Its time-aware shaper protocol tackles uncertain delay and frame loss but introduces high scheduling complexity. However, existing works lack a scheduling feature mining mechanism. They impose unnecessarily tight rules to simplify the problem but sacrifice scheduling optimality. To address this, especially in industrial networks with large-scale complex flows, we propose a flow-overlap graph based distributed scheme to improve scheduling scalability concerning schedulability, scheduling efficiency, and latency and jitter. The distributed framework is established with the pipeline-parallelism pattern and verified superior in scalability. It first incorporates the deterministic feature into the distributed TSN configuration standard. Under this, specific scheduling is refined by building a so-called flow-overlap graph that efficiently characterizes flow-based scheduling features and further designing a hierarchical scheduling algorithm GFD. This scheme Pareto dominates the three scalability criteria theoretically and simulatively.

The sequential decision-making problem with large-scale state spaces is an important and challenging topic for multitask reinforcement learning (MTRL). Training near-optimality policies across tasks suffers from prior knowledge deficiency in discrete-time nonlinear environment, especially for continuous task variations, requiring scalability approaches to transfer prior knowledge among new tasks when considering large number of tasks. This paper proposes a multitask policy adversarial learning (MTPAL) method for learning a nonlinear feedback policy that generalizes across multiple tasks, making cognizance ability of robot much closer to human-level decision making. The key idea is to construct a parametrized policy model directly from large high-dimensional observations by deep function approximators, and then train optimal of sequential decision policy for each new task by an adversarial process, in which simultaneously two models are trained: a multitask policy generator transforms samples drawn from a prior distribution into samples from a complex data distribution with higher dimensionality, and a multitask policy discriminator decides whether the given sample is prior distribution from human-level empirically derived or from the generator. All the related human-level empirically derived are integrated into the sequential decision policy, transferring human-level policy at every layer in a deep policy network. Extensive experimental testing result of four different WeiChai Power manufacturing data sets shows that our approach can surpass human performance simultaneously from cart-pole to production assembly control.

Large-scale machine coordination is a primary approach for flexible manufacturing, enabling large-scale autonomous machines to dynamically coordinate their actions in pursuit of a custom task. One of the key challenges for such large-scale systems is finding high-dimensional coordination decision-making policies. Multitask policy gradient algorithms can be used in search of high-dimensional policies, particularly in collaborative decision support systems and distributed control systems. However, it is difficult for these algorithms to learn online high-dimensional coordination control policies (CCP) from large-scale custom manufacturing tasks. This paper proposes a large-scale online multitask learning and decision-making approach, which can consecutively learn high-dimensional CCP in order to quickly coordinate machine actions online for large-scale custom manufacturing task. A large-scale online multitask leaning algorithm is developed, which is able to learn large-scale high-dimensional CCP in a flexible manufacturing scenario. An online stochastic planning algorithm is proposed, which online optimizes the Markov network structure in order to avoid expensive global search for the optimal policy. Experiments have been undertaken using a professional flexible manufacturing testbed deployed within a smart factory of Weichai Power in China. Results show the proposed approach to be more efficient when compared with previous works.