Interoperability and Interworking of Intelligent Transportation Systems and the Rise of AI

Introduction

The United Nations Economic Commission for Europe (UNECE) has invited me back in 2018, to develop a case study on standards and their role in achieving the Sustainable Development Goals (SDGs). My first choice was to develop a study on Intelligent Transportation Systems (ITS) systems and services and their key role in achieving SDG11.

ITS involve vehicles, drivers, passengers, road operators, and city managers all interacting with each other and with the environment, linking with often complex backbone infrastructure systems. One of the primary targets of ITS is to reduce the number of traffic accidents, as well as improving the efficiency of traffic networks. Sustainability is an equally important dimension and there is currently a deep focus on the impact of transport systems on the environment.

Furthermore, with the widespread adoption of electric vehicles worldwide, and the trend of incorporating Artificial Intelligence (AI) systems in transportation systems and the rise of autonomous vehicles, it is argued that such use of increasingly autonomous vehicles should be guided by the policy of human control, according to which humans should execute a certain significant level of judgment over such systems.

ITS systems are by their very nature, interactive. To operate successfully, data must be sent accurately and in a timely manner. Furthermore, data must find the correct recipient and be understood by that recipient, who may be in a completely different system. Such systems can only interoperate and interwork successfully if they are designed to comply with international standards.

In this article intends to re-explore parts of the previous study delivered to the UNECE, on the interoperability of ITS services and systems using standards and investigate how AI has changed and/or is foreseen to change the standardization landscape of this complex and evolving ecosystem.

Interoperability and Interworking of ITS System Components

ITS entail the integration of many different technologies, and systems from many vendors and technology providers. Typical objectives include increasing road transportation safety and security along with enhancing mobility and transport efficiency. This in turn requires the design and implementation of many integrated modules. Examples include enforcement applications (e.g. speeds detectors systems, weigh in motion, and other traffic violations detection), tolling, passenger information systems, and traffic management.

There are different approaches of integration depending on the required level of monitoring and control. Options include integration at the field devices level, or the aggregation nodes, or at the control centre levels or a hybrid approach. Adopting an open architecture and selecting an optimal integration point is the recommended approach.

With the rise of autonomous vehicles, smart and efficient integration become more pressing. In such cases, where the vehicle is no longer being driven by humans, but rather by an AI system which controls the vehicle, and its subsystem; it is imperative to address interoperability in the most effective and efficient manner. Relying on a multitude of sensors, AI systems controlling the vehicle’s anti-collision and emergency systems must communicate with road-side equipment, and/or other ITS infrastructure, in a split of a second.

Vehicle manufacturers are constantly equipping their vehicles with driver assistance and support systems, which already constitute some sort of partial automation systems. The American Society of Automotive Engineering (SAE) has set a classification of vehicle autonomy levels, the SAE J3016, by defining the SAE Levels from Level 0 (no driving automation) to Level 5 (full driving automation) in the context of motor vehicles and their operation. Higher autonomy levels mean greater AI complexity, as well as intensive use of sensors in the vehicle. Different autonomy levels demonstrate different driver level involvement and different complexities of the AI system controlling the vehicle. Such taxonomy in the level of vehicle “smartness” from a standardization point of view requires further consideration across Standards Development Organizations (SDOs) through a genuine collaboration and cross-industry engagement efforts.

Indeed, autonomous vehicles have control systems that detect and respond to events in their vicinity. However, this shouldn’t be mistakenly considered, as a bullet-proof vest. There are intrinsic limitations for such control system, specifically due to the way the sensory data are processed and interpreted. For example, there are wide variations in the vehicle’s surrounding environment, which include varying weather and lighting conditions, dynamic traffic conditions, and unpredictable driving pattern of the vehicle’s driver.

Interoperability and interworking with smart city systems and services, needs a re-engineering approach, whereby modifications in the ITS protocols stack and interfaces are envisioned to allow the vehicles (regardless of their SAE automation level) and the ITS infrastructure elements, to communicate and exchange data and information in a reliable, secured, and fast manner.

Strategy: Implementing an AI-Enabled ITS Standardized Architecture

Implementing a standardized architecture facilitates the process of laying down the required functionalities for any system to operate effectively and efficiently. Using a standardized communication infrastructure enables the integration of different equipment from different vendors. The communication infrastructure can use a wide range of standards depending on requirements analysis.

For example, using include standardized optical fiber infrastructure (e.g. by implementing ITU-T standardized G Series recommendations) to connect the road-side equipment and other mission critical ITS infrastructure components to an intelligent backend whereby all infrastructure components and the vehicle cooperate (e.g. by implementing ITU-T standardized Y, H, and X Series recommendations) to avoid any foreseen collision in a split of a second, by deploying low latency sensory data fusion systems; extremely high speed and reliable secured Field-to-Traffic Management Center communications (Field-to-Center) and Traffic Management Center-to-Field communications (Center-to-Field); AI-based data processing; and intelligent control of field equipment and on-board vehicle devices (depending on the SAE vehicle autonomy level).

Specialized urban traffic controllers which follow some specifications like Open Communication Interface for Road Traffic Control Systems (OCIT), Sydney Coordinated Adaptive Traffic System (SCATS), Split Cycle Offset Optimisation Technique (SCOOT), can be difficult to interwork with an AI-controlled vehicle, or communicate with any third-party smart city infrastructure element, given their legacy nature. However, given its wide-spread use, the National Transportation Communications for ITS (NTCIP) protocols can be used as a foundation, with some modifications, to connect field devices in a standardized manner.

NTCIP devices are not in general compatible with some widely adopted urban traffic controllers, and this complicates the integration process a little bit. However, specific measures are needed to aggregate and process the data at the device level in the field, or at the edge, and/or back-end levels to overcome these challenges.

High Safety Assurance Zones

Having a city-wide vehicle safety assurance framework require continuous or quasi-continuous communications with the vehicles. This might be difficult to realize given the heterogeneity of the infrastructures deployed in any particular city. However, designing a “High Safety Assurance Zone” can be a more feasible target.

In a “High Safety Assurance Zone”, both the vehicle and the smart city infrastructure (e.g. urban traffic controllers, road-side equipment, gantry-mounted anti-collision systems, intelligent speed bumps, variable message signs…etc.) can all work together in cooperative sensing, modeling, and control to avoid possible collisions, and to reduce the likelihood of accidents.

Furthermore, smart city ITS infrastructure can be equipped with the adequate systems to communicate with interoperable smart autonomous vehicles, to control the vehicle and/or the traffic, to reduce the probability of collisions.

The whole “Zone” is cooperating to achieve one target:

“The safe delivery of the vehicle’s passengers and/or package or freight to their intended destinations in the most effective, secured and efficient way”

Think of a scenario whereby a gantry equipped with an intelligent anti-collision system, detected a possible collision involving an incoming autonomous vehicle (based on its trajectory/speed and the trajectory/speed of other close by objects) and decided to engage its automatic brakes to avoid the collision.

One notable effort to achieve this vision, is what the ITU-T is currently developing in its draft Recommendation Y.IoT-IWAT “Framework of interworking with agent-based transportation for intelligent IoT services”. This work is being developed by industry experts in ITU-T SG20, the key expert group developing standards on IoT, digital twins, and smart sustainable cities and communities. Figure 1 presents an example of interworking with agent-based transportation for intelligent IoT services, from this draft Recommendation.

The industry is yearning for standards that would unify, or better interoperate and interwork different systems together. The proliferation of IoT and other commonly used IoT platforms make things even more complicated. Efforts to standardize IoT platforms, which can also serve ITS applications, can be found in ITU-T Y.4200 and Y.4201 which developed critical requirements for the interoperability of smart city platforms, along with the requirements and reference framework of smart city platforms, respectively.

The Role of Standards

Standards play a key role in interoperability and interworking of different vehicle and smart cities/ITS components and systems, avoiding vendors lock-in, by ensuring global competitiveness, and hence achieving better quality of services, and higher performance versus costs options. Moreover, standards play a crucial role in ensuring the future system’s scalability, by adopting a modular approach based on a standardized architecture. They lower the overall total cost of ownership (TCO) and provide some level of guarantee that the technology being developed represents various industry views and various stakeholder’s segments.

Challenges and Lessons Learned

In the author views, the main challenges ahead, is still in the difficulty of selecting the standards at the different levels of the system. Standards themselves, especially on regional levels entail differences which are triggered in principle by competing industry ecosystems. This highlights the importance of international standards developing organizations (SDOs) like the ITU, IEC, ISO to develop an internationalized set of standards or at least principles that would unify requirements and/or architectures to realize global cross-industry interoperability and interworking.

Interoperability and interworking aspects are becoming more difficult with the proliferation of new and emerging technologies like the AI, big data, and cloud. However, designing and implementing “High Safety Assurance Zones” as a confined space, presents a plausible and an achievable target. There is a need to engage all relevant stakeholders on a wider scale, experts from the academia, industry , govern, and civil society, to drive the standardization ecosystem beyond its traditional boundaries, exploring new avenues of technology development at the intersection of transportation and smart city infrastructure, cybersecurity, products safety, quality assurance, AI, IoT, and smart sustainable cities and communities.

Join us at the ITU and the U4SSC to shape the future of AI, IoT, Digital Twins, and Smart Sustainable Cities and Communities.

About the Author

Ramy Ahmed Fathy, PhD

Vice-Chair of ITU-T Study Group 20 (IoT, Digital Twins, and Smart Sustainable Cities and Communities

Vice-Chairman of United for Smart Sustainable Cities (U4SSC)

The United for Smart Sustainable Cities (U4SSC) initiative is a global UN collaboration, coordinated by ITU, UNEP and UNECE, and supported by a network of key partners, including UN-Habitat, CBD, ECLAC, FAO, UNDESA, UNDP, UNECA, UNESCO, UNEP, UNEP-FI, UNFCCC, UNIDO, UNOPS, UNU-EGOV, UN-Women, UNWTO, and WMO. U4SSC serves as an international platform for exchanging knowledge and fostering partnerships to empower cities and communities in achieving the UN Sustainable Development Goals. By engaging with its expert membership in key thematic groups, the U4SSC expert develop actionable plans, technical specifications, case studies, and guidelines, providing cities with the tools and policy guidance they need to enhance their sustainability and accelerate their digital transformation. By offering tailored solutions and insights, U4SSC helps cities around the world become more efficient, innovative, and people-centered—paving the way for a smarter, greener future.

?

?