Navigating Tomorrow:​ Urban Mobility's Future Challenges​

The lack of direct line-of-sight (LOS) signals from mainstream Global Navigation Satellite Systems (GNSS) in densely built urban canyon environments presents a significant challenge to achieving precise Positioning, Navigation, and Timing (PNT) accuracy. Despite the existence of several available techniques and solutions, the current state of PNT accuracy falls far short of what is achievable in non-urban canyon environments. This limitation becomes particularly critical with the rise of new applications such as autonomous driving and indoor positioning, where accurate location information is paramount.

In response to this challenge, emerging complementary technologies offer promising avenues for addressing the deficiencies of GNSS in urban canyon environments. Among these technologies are Low Earth Orbit (LEO) satellite constellations, terrestrial networks like 5G and the anticipated 6G, as well as visual sensors. By leveraging a combination of these technologies, it becomes possible to obtain accurate location information both outdoors and indoors, overcoming the limitations posed by urban canyon environments. Integration of LEO satellite constellations can provide additional satellite coverage with improved visibility in urban canyons, mitigating the effects of signal blockage and multipath interference experienced with traditional GNSS systems. Terrestrial networks like 5G and 6G offer the potential for precise positioning through advanced signal processing techniques and dense network infrastructure deployment, further enhancing location accuracy in urban environments.

Visual sensors, such as cameras and LiDAR (Light Detection and Ranging) systems, offer an alternative or complementary approach to satellite-based positioning technologies. By capturing and processing visual data from the surrounding environment, these sensors can provide accurate location information independent of satellite signals, making them well-suited for indoor environments or areas with limited GNSS visibility. To achieve robust and accurate location determination in urban environments, it is imperative to integrate and synergize these diverse technologies and solutions.

To ensure the safety of AI systems, particularly deployed within autonomous vehicles operating in urban environments, developers have increasingly turned to black-box validation algorithms. These algorithms are characterized by their inherent opacity, whereby users can only input data and observe the ensuing outputs, without direct insight into the internal mechanisms governing these outcomes. This approach allows for rigorous testing and validation of the AI systems' behaviour across various scenarios, contributing to enhanced safety and reliability in complex urban environments. Given the criticality of safe algorithms in accurately perceiving and reacting to the vehicle's surroundings, thereby safeguarding passengers and other road users, meticulous testing and verification are paramount.

The inherent complexity of these systems and the intricacies of their operational environments necessitate rigorous validation protocols. While road tests are indispensable in assessing vehicle safety, they are typically conducted in the latter stages of the design process and entail significant risks. Consequently, engineers are compelled to devise methods for validating autonomous vehicle algorithms pre-emptively, prior to on-road experimentation.

Ensuring compliance with stringent safety standards mandated by organizations such as the International Organization for Standardization (ISO) and the Society of Automotive Engineers (SAE) is imperative. Notably, ISO 26262, a widely embraced functional safety standard, plays a pivotal role in guaranteeing the secure functioning of the electrical and electronic systems within road vehicles.

Drones are increasingly becoming a pivotal component of urban mobility, providing essential services such as delivery, aerial surveying, infrastructure inspection, and emergency response. In dense urban environments, the accuracy and reliability of PNT systems are paramount for the safe and efficient operation of drones. These environments, characterized by tall buildings and narrow streets, present unique challenges for PNT, similar to those faced by autonomous vehicles. Traditional GNSS often suffer from signal blockages, multipath interference, and degraded accuracy in these urban canyons.

To overcome these challenges, a multifaceted approach integrating various technologies and methodologies is required. Low Earth Orbit (LEO) satellite constellations can offer enhanced coverage and mitigate GNSS limitations. Terrestrial networks such as 5G and the forthcoming 6G provide high-precision positioning through dense infrastructure and advanced signal processing. Additionally, visual sensors, including cameras and LiDAR, enable precise navigation by capturing environmental data independent of satellite signals. These technologies, when combined through sensor fusion techniques, can significantly enhance the PNT capabilities for drones operating in urban settings.

Regulatory issues and the importance of standardization are critical factors in the deployment of PNT-enabled drone solutions in urban areas. The lack of standardized protocols and regulations can lead to inconsistencies in system performance and safety, hindering widespread adoption. Regulatory frameworks need to address concerns related to privacy, airspace management, and the integration of drones into existing urban infrastructure. Standardization efforts are essential to ensure interoperability between different PNT technologies and to establish uniform safety and performance benchmarks. Moreover, regulatory bodies must collaborate with industry stakeholders to create comprehensive guidelines that promote innovation while safeguarding public safety and privacy. Addressing these challenges requires a concerted effort to harmonize regulations across regions and develop robust standards that support the seamless operation of drones in dense urban environments.
 

As advancements in technology continue, PNT technologies including Global Navigation Satellite Systems (GNSS), are increasingly becoming integral components of autonomous systems across various automation levels, such as self-driving vehicles, drones, and highway lane-keeping systems. Despite their advantages, different PNT systems pose different risks. In particular GNSS signals are weak and easily interfered with or blocked. Intentional interferences of PNT signal include Jamming and Spoofing. Jamming involves disrupting communications by transmitting electromagnetic energy in the same radio frequency bands as the targeted signal. Spoofing, on the other hand, mimics GNSS signal structure and content to deceive receivers into generating false positions, posing a greater threat as the receiver is unaware of the attack. Numerous incidents of GNSS jamming and spoofing have been reported globally, highlighting the potential risks associated with these forms of interference. Considering the vital function of PNT information in autonomous systems, where it aids navigation and decision-making, any failure in PNT signals could significantly impair the proper operation of these systems. Consequently, such failures would elevate risks not only for the vehicle's driver but also for other traffic participants. In addition, ensuring the security of overall communication, is imperative for establishing trust in the vehicle's actions. Securing the transmission and reception of vehicle information is essential to guaranteeing the integrity and reliability of the vehicle's navigation and decision-making processes. Without robust security measures in place, there is a heightened risk of malicious interference or manipulation, which could compromise the safety of the vehicle and its occupants. Therefore, bolstering the security of communication channels for transmitting PNT data is paramount for instilling confidence in the functionality and behaviour of autonomous vehicles. 

In recent decades, numerous indoor and outdoor applications of wireless communication have spread rapidly. One critical aspect of this expansion is the emergence of device-to-device (D2D) communication, which facilitates connectivity among various types of networks and enables communication between devices. This has particularly flourished in conjunction with the advancements in 5G and beyond, fostering flexible networking capabilities among interconnected devices. Vehicular Ad Hoc Networks exemplify this trend. Leveraging the framework provided by the IEEE 802.11 networking standards, communication with and between vehicles has been realized, enhancing on-road safety and comfort through real-time environmental awareness and traffic monitoring. Initially, the integration of low-cost GNSS and Wireless Local Area Network (WLAN) technologies aimed to address key objectives: enhancing road safety, reducing environmental impact associated with transportation, and optimizing traffic flow. This translates into tangible benefits such as decreased traffic accidents, resulting in fewer on-road congestion instances, thereby improving travel quality through reduced commute times, and mitigated environmental repercussions. While the general objectives appear clear, the implementation and application are still lacking, in particular for urban areas. One reason is the missing capabilities to integrate applications and technologies directly into the system. In this context, leveraging Intelligent Digital Twins (IDT) to enhance Software-Defined Vehicular Networks (SDVN) can play a crucial role, as PNT information can significantly contribute to constructing a realistic and dynamic virtual representation of the physical network environment within the IDT.

This data allows for simulating traffic patterns, modelling the movement of vehicles and network entities, and characterizing the physical environment, ultimately leading to more accurate predictions and verifications of network performance. Furthermore, real-time PNT data can validate the accuracy of the virtual network model, identify discrepancies between the real and virtual environments, and trigger updates to the networking schemes based on real-world changes. By providing these functionalities, PNT acts as a critical bridge between the physical and virtual network environments, ensuring the effectiveness and adaptability of IDT-SDVN in optimizing vehicular networking. This approach can also be used to assess the performance and responsiveness of UAVS in urban areas. 

The link between integrated Traffic Management (TM 2.0 concept and Mobility as a Service (MAAS) has been progressively built during the past five years as a pathway towards addressing the complex transportation needs of densely populated urban areas while keeping the mobility network less congested and operational. Traffic Management operations has long realised that only focussing on car traffic is not enough when other road-based modes (such as buses, trams, bicycles and mopeds) are equally affecting and interacting with traffic infrastructure (such as traffic lights, crossings etc.). On the other hand, by integrating various transportation modes into a seamless and interconnected system, Mobility as a Service (MAAS) claims to offer an efficient and sustainable solution to urban mobility needs of users. Nonetheless, the successful implementation of both the Multi-Modal Mobility Management and MAAS relies heavily on reliable PNT information as it enables accurate and real-time location-based services across different transportation modes. Whether it's navigating through congested city streets, coordinating transfers between buses and trains, or hailing a ride-sharing service, PNT technologies provide the necessary spatial awareness to optimize travel routes, minimize delays, and enhance the overall user experience. Both users and traffic management service providers are in need of reliable quality PNT information. In dense urban environments, where tall buildings, narrow streets, and signal obstructions are prevalent, ensuring the accuracy and reliability of PNT data poses significant challenges. Multipath interference, signal attenuation, and urban canyon effects can degrade the performance of traditional GNSS, necessitating the development of alternative PNT solutions tailored to urban environments. Furthermore, the integration of PNT technologies with smart city infrastructure is essential for optimizing traffic flow, reducing congestion, and improving overall transportation system efficiency. By leveraging real-time data from sensors, connected vehicles, and intelligent transportation management systems, both Multimodal management and MAAS platforms can offer dynamic trip planning capabilities and personalized route recommendations tailored to individual preferences and constraints as well as public authority priorities. One of the challenges in Multimodal mobility management and MAAS lie in the data privacy and security concerns, interoperability and integration issues, cost and scalability constraints, and regulatory and policy considerations. Addressing these challenges requires collaboration between stakeholders across the public and private sectors to develop robust a PNT infrastructure, establish clear guidelines and standards, and ensure compliance with relevant laws and regulations.