The challenge of ensuring safety in autonomous driving or sailing involves predicting and replicating various potential scenarios on roads and waterways, posing difficulties and high costs. In response, the European project 5G-Blueprint addresses this by introducing a complementary technology, i.e., teleoperation, which leverages 5G connectivity to enable human interventions in complex situations beyond autonomous capabilities, thereby removing the physical link between the human operator and the remotely controlled vehicle/vessel. This operational mode brings stringent connectivity requirements, including high uplink bandwidth for transmitting video streams from onboard cameras to the teleoperation center, low latency, and an ultra-reliable connection for relaying commands from the teleoperator to the remote vehicle/vessel. Additionally, it emphasizes minimal interruption time when the teleoperated vehicle/vessel crosses international borders, ensuring seamless connectivity and uninterrupted remote operation. Therefore, this paper summarizes extensive evaluations of network and service performance, highlighting key results across pilot locations and providing conclusions and analysis of 5G-enhanced teleoperation in various use cases. Additionally, it outlines lessons learned from pilot activities.

The International Transport Forum (ITF) predicts a significant increase in demand for transportation in the coming years, despite the shortage of drivers. To tackle this challenge, the Transport and Logistics (T&L) industry is increasingly relying on emerging technologies. While connected and autonomous driving offer promises of greater safety, efficiency, and environmental benefits, connected and autonomous driving face operational hurdles in complex environments. However, the existing limitations of autonomous vehicles, particularly in dense urban settings, highlight the need for complementary technologies, such as teleoperation. The European Horizon 2020 5G-Blueprint project aims to design and validate the technical architecture and business models for cross-border teleoperated transport, utilizing 5G technology. This study delves into the implementation of a real 5G Standalone (5G SA) network within a port environment, utilizing network slicing for teleoperation and Multi-Access Edge Computing (MEC) to enable real-time video processing at the network edge. Specifically focusing on Ultra-Reliable Low Latency Communications (URLLC) and enhanced Mobile Broadband (eMBB) slices, we conduct a comprehensive evaluation of a real-world 5G SA network. Our assessment examines key performance parameters such as Round-Trip Time(RTT) latency, Packet Delivery Rate (PDR), Reference Signals Received Power (RSRP), and corrupted frame rates, emphasizing the crucial role of 5G network slicing and MEC in enhancing operational reliability and efficiency in teleoperated transport systems.

The Transport and Logistics (T&L) sector faces numerous challenges, including the search for qualified personnel, as well as improving driver safety and work-life balance. Teleoperation emerges as the technology able to address these challenges. Thanks to 5G connectivity and network slicing, operating vehicles remotely from a Teleoperation Center (ToC) is becoming a reality. The European project 5G-Blueprint, funded by the European Union, has demonstrated the feasibility of 5G-based teleoperation, even in a cross-border context. Despite the fact that 5G and network slicing enable reliable and low-latency transmission of video data from cameras installed on Teleoperated Vehicles (ToVs) to ToC, the perception of the surrounding environment is different for the teleoperator compared to the driver who is physically present in the vehicle. In this paper, we introduce a real-world system that showcases synergy among different teleoperation elements, including intelligent traffic lights (iTL) and Vulnerable Road Users (VRU), aimed at supporting teleoperation by improving remote driver’s situational awareness. This synergy enhances the environmental perception of the teleoperator, bridging the gap between their experience and that of an in-vehicle driver. First, we evaluate the performance of a real-world 5G network with network slicing, based on actual data and testing scenarios conducted in both industrial and urban areas with 5G Standalone (5G SA) coverage. Then we validate the 5G capabilities for enabling a real-world system that showcases synergy among different teleoperation elements.

The native integration of AI and ML algorithms in the next-generation mobile network architecture will allow for meeting the expectations of 6G. This aspect is targeted by the DAEMON project, which proposed a solution to natively manage Network Intelligence (NI) through novel architectural elements and procedures. In this paper, we discuss how NI solutions based on AI and ML can leverage NI native procedures implemented by the NI Orchestrator to improve their lifecycle management. We also discuss how the architectural procedures can be implemented in practice, using state-of-the-art software components.

Network slicing enables multiple virtual networks to share physical resources, allowing network operators to deliver highly customizable and efficient networking solutions that meet the diverse requirements of modern applications. The automated management of network slices has been studied in the last years to make such solutions more flexible, ready to support new applications, and capable of optimizing network resource utilization. Many works in the literature give a top-down approach, focusing on the high-level decision processes, and relying on abstracted infrastructure managers and simulation tools to apply/execute such decisions. In this work, we leverage components that we previously developed for network monitoring, flexible traffic shaping, and Software-Defined Time-Sensitive Networking control, to create a bottom-up approach toward automated slice management. We describe the intricate coordination of elements required for an automated control loop and present the results achieved with a proof-of-concept executed in a real testbed of wired and Wi-Fi nodes. The results show the capability of the system to correctly identify the bottleneck of a flow and apply corrective actions to reestablish its intended performance level.

Intent-driven network management has become an important part of autonomous systems in Beyond 5G (B5G) towards Sixth-Generation (6G) networks, by enabling flexibility in the interaction among applications, operators and users. Intents play an important role in the communication of road users like autonomous vehicles and pedestrians to edge computing services. As sensor technologies for modern vehicles are cheaper, smaller, diverse and computing capable, more demand for applications and services on the road is increasing. A flexible intent interpretation and coordination are needed to deal with the dynamic environment and constantly changing goals. This paper presents a proof-of-concept of Zero-touch Network and Service Management (ZSM) for vehicular communication services, using an Intent Management Entity (IME) to translate user objectives into actionable directives. This paper describes a realistic testbed setup at the Smart Highway, where a Deep Reinforcement Learning (DRL) algorithm is used to optimize the selection of Roadside Units (RSUs) for service orchestration. This paper also discusses the challenges and opportunities of enhancing the IME with time-based intent coordination, using Artificial Intelligence and Machine Learning (AI/ML) techniques to estimate the waiting time and priority in intent coordination. The paper aims to demonstrate the benefits of ZSM and Intent-driven Management for vehicular edge computing and B5G/6G autonomous network management frameworks.

In this paper, we demonstrate and introduce a novel Situational Awareness with Event-driven Network Programming Edge Network Application (EdgeApp), designed to optimize network resource utilization during vessel teleoperation in congested port areas. The demonstration is conducted on an open real-life EdgeApp 5G Standalone (SA) and beyond testbed situated at the port of Antwerp-Bruges. Through this showcase, we demonstrate how 5G and beyond services, utilizing an open 5G SA testbed, can enhance vessel teleoperation. The proposed solution dynamically adjusts network configurations, allowing for lower-quality camera feeds during vessel autonomy and higher-quality feeds when in the teleoperation zone. The practical application and benefits are exemplified through visual representations within the testbed environment.

The evolving landscape of 5G Standalone (SA) and beyond networks is being increasingly focused on vertical industries. To unlock the full potential for verticals, it is important to tightly integrate Edge Network Applications (EdgeApps) tailored to vertical use cases, with the 5G SA network, while allowing them to interact with each other at the same time. Such interaction enables more transparency in expressing Quality of Service (QoS) demands from verticals in the form of intent while hiding the network complexity from them. In this paper, we propose two EdgeApps, which by interacting with both User Equipments (UEs) and 5G SA network, are becoming aware of network quality (quality-awareness) and context around UEs (situational-awareness). Such awareness is also enabled by initiatives such as GSMA Open Gateway and CAMARA, where network and IT functionality are exposed to application developers through standardized Application Programming Interfaces (APIs) that abstract the underlying complexity on the telco and IT systems. In this paper, we utilize the Nokia Network as Code (NaC) platform that exposes the capabilities of the Telenet 5G SA network through CAMARA APIs allowing our EdgeApps to dynamically and in real-time create events that trigger changes in QoS levels required by vertical applications. The paper showcases this concept through a case study within the Transport and Logistics (T&L) sector, which is focused on improving the safety and efficiency of remote vessel operation in busy port environments. The overall solution is deployed and tested on the Antwerp 5G SA testbed, which consists of the UEs (vehicle and vessel) and 5G network infrastructure in the Port of Antwerp-Bruges, as well as the 5G edge where EdgeApps are running. This research contributes to the broader objective of incorporating diverse industrial applications into the 5G and beyond ecosystem, showcasing tangible benefits for vertical industries.

Automating network processes without human intervention is crucial for the complex 6G environment. This requires zero-touch management and orchestration, the integration of Network Intelligence (NI) into the network architecture, and the efficient lifecycle management of intelligent functions. Reinforcement Learning (RL) plays a key role in this context, offering intelligent decision-making capabilities suited to networks' dynamic nature. Despite its potential, integrating RL poses challenges in model development and application. To tackle those issues, we delve into designing, developing, and validating RL algorithms for scaling network functions in service-based network architectures such as Open Radio Access Network (O-RAN). It builds upon and expands previous research on RL lifecycle management by proposing several RL algorithms and Reward Functions (RFns). Our proposed methodology is anchored on a dual approach: firstly, it evaluates the training performance of these algorithms under varying RFns, and secondly, it validates their performance after being trained to discern the practical applicability in real-world settings. We show that, despite significant progress, the development stage of RL techniques for networking applications, particularly in scaling scenarios, still leaves room for significant improvements. This study underscores the importance of ongoing research and development to enhance the practicality and resilience of RL techniques in real-world networking environments.

As network complexity escalates, there is an increasing need for more sophisticated methods to manage and operate these networks, focusing on enhancing efficiency, reliability, and security. A wide range of Artificial Intelligence (AI)/Machine Learning (ML) models are being developed in response. These models are pivotal in automating decision-making, conducting predictive analyses, managing networks proactively, enhancing security, and optimizing network performance. They are foundational in shaping the future of networks, collectively forming what is known as Network Intelligence (NI). Prominent Standard-Defining Organizations (SDOs) are integrating NI into future network architectures, particularly emphasizing the closed-loop approach. However, existing methods for seamlessly integrating NI into network architectures are not yet fully effective. This paper introduces an in-depth architectural design for a Network Intelligence Stratum (NI Stratum). This stratum is supported by a novel end-to-end NI orchestrator that supports closed-loop NI operations across various network domains. The primary goal of this design is to streamline the deployment and coordination of NI throughout the entire network infrastructure, tackling issues related to scalability, conflict resolution, and effective data management. We detail exhaustive workflows for managing the NI lifecycle and demonstrate a reference implementation of the NI Stratum, focusing on its compatibility and integration with current network systems and open-source platforms such as Kubernetes and Kubeflow, as well as on its validation on real-world environments. The paper also outlines major challenges and open issues in deploying and managing NI.