1
Computational Intelligence and Internet of Vehicle Things: An Introduction

Akshya. J.1*, Sundarrajan M.2, Mani Deepak Choudhry3, Kirubhakaran M.4 and Reeba Rose L.5

1Department of Computational Intelligence, School of Computing, SRM Institute of Science and Technology, Kattankulathur, India

2Department of Networking and Communications, School of Computing, SRM Institute of Science and Technology, Kattankulathur, India

3Department of Computing Technologies, School of Computing, SRM Institute of Science and Technology, Kattankulathur, India

4Department of Computer Science and Engineering, Surya Engineering College, Erode, India

5Department of Computer Science and Engineering, St. Joseph's College of Engineering, Chennai, India

Abstract

Computational intelligence (CI) and the Internet of Vehicle Things (IoVT) are reshaping the paradigm of intelligent transportation, and their interaction is influencing the paradigm shift of the decision process in transportation by elevating traditional vehicular networks to decentralized, adaptive, and cognitively accustomed mobility systems. The result of such integration is that vehicles can not only act as ferrying entities, but they would also be able to gauge, reason, and interact in dynamic ways within the complexity of real-time scenarios. In comparison to conventional rules-based solutions, however, CI techniques, such as deep learning, fuzzy logic, reinforcement learning, and evolutionary algorithms, enable IoVT systems to provide predictive maintenance, adaptive cruise control, context-aware navigation, real-time traffic optimization, and more robust cybersecurity. The CI-enabled Internet of Virtual Things architecture is horizontally distributed over edge, fog, and clouds layers, whereby decisions are made quickly, data is processed in real time, and the overall coordination across the entire system is scalable. Equipping vehicles and infrastructure with intelligence allows such systems to serve such crucial roles as collision avoidance, swarm synchronization, and anomaly detection, and become highly resilient to cyber threats by living in a federated and adaptive trust version. The issues of scalability, standardization, ethical AI governance, and responsiveness in real-time, however, still have yet to be addressed, which requires researcher-policymaker-industry collaboration across sectors. The vision ahead in autonomous vehicle networks is an interconnected, self-organizing, and ethically compatible global system of vehicles, drones, and pedestrians all working in harmony through the orchestration of CI. Such networks will learn, adapt, and evolve in a continuous way so that smart cities will be safer, greener, and more inclusive. This incorporation of CI into IoVT is not only a technological improvement but also a shift in paradigm towards an intelligent mobility infrastructure that is tailored to the rest of the society with the view of achieving its various integrities in terms of sustainability, safety, and digitalization.

Keywords: Computational intelligence, Internet of Vehicle Things, autonomous vehicles, smart transportation, intelligent mobility systems

1.1 Introduction to Computational Intelligence in Internet of Vehicle Things

The arrival of Internet of Vehicle Things (IoVT) is a new era for transportation, where vehicles are not isolated mechanical units, but rather become intelligent nodes within a broad-acre digital system. It extends the concepts of the traditional Internet of Things by providing a real-time data exchange, making dynamic decisions, and enabling autonomous vehicle network operations. In this landscape, CI has become one of the key pillars that allow vehicles to concurrently perceive, reason, and adapt to the environment dynamically. Methods such as fuzzy logic, application of neural networks, use of evolutionary algorithms, and hybrid models are included in CI, which incorporates application capabilities to allow the systems to learn from complex, imprecise, or incomplete information. CI techniques are useful in the context of IoVT to provide predictive maintenance, context-aware navigation, intelligent traffic routing, accident prediction, and autonomous decision-making ability. As opposed to rigid rule-based systems, computational intelligence introduces flexibility in traffic conditions when they are uncertain and dynamic, thus providing robustness and tolerance to faults. In addition, the introduction of CI into IoVT is essential to cope with the huge quantities of distinct data produced by vehicles and infrastructure systems. Using methods like deep learning and reinforcement learning, the vehicles can find inherent patterns in the flow, anticipate future traffic, and take preventive measures so that it does not face any risks. It is not only technological advancement but also a move to advancing vehicles to operate as cognitive vehicular ecosystems, vehicles that think, learn, and work together autonomously. CI in IoVT has amalgamated a new level of intelligence, resilience, and efficiency to vehicular systems and redefined transportation paradigms. With the convergence of science, industry, and research institutions in the direction of developing smart transportation infrastructures, the deployment of computational intelligence not only makes sense but also is absolutely mandatory for the future of the next generation of connected mobility systems.

Figure 1.1 shows the structure of intelligent vehicle networks that comprises four important communication modes: vehicle to vehicle (V2V), vehicle to infrastructure (V2I), vehicle to pedestrian (V2P), and vehicle to network (V2N). Through establishing these modes, vehicles and nascent pods, pedestrians, pedestrian office units, and network systems, which are the basis of the modern transportation ecosystems, communicate seamlessly with each other. These interactions are connected to three areas that enable the development: smart transportation systems which optimize the flow and traffic safety, computational intelligence that permits decision-making in real time and anomaly detection, and 5G and the emerging 6G technologies, on which this all relies upon the ultra-reliable and low latency communication infrastructure required for efficient and autonomous vehicular operation. It is the driving force of the intelligent mobility future.

Figure 1.1 Intelligent vehicle networks.

1.1.1 Emergence of Intelligent Vehicle Networks

Intelligent vehicle network emerges as a huge revolution from the basic vehicular communication system to a highly dynamic, decentralized, and self-organized ecosystem. Vehicular ad-hoc networks (VANETs) or connected vehicle systems (CVSs) are typically intelligent vehicle networks characterized in terms of their ability to allow vehicles (V2V), vehicles and infrastructure (V2I), vehicles and pedestrians (V2P), and vehicles and network (V2N) to communicate. Using these networks, smart transportation systems will be enabled with the ability to broadcast, disseminate real-time data in a networked fashion, and coordinate data with cooperative awareness, as well as coordinated decisions by one or more vehicles. Over the past decades, with the increasing urgency for improved road safety, higher traffic efficiency, better environmental sustainability, and better user experience, intelligent vehicle networks have driven the development. But they are not passive subjects of control commands anymore; they are now active participant in intelligence assembly, able to sense the context, translate the data by their own intelligence, and make autonomous decisions based on the collective intelligence. The prediction of the behavior of surrounding entities and the vehicle's ability to adaptively react to changing conditions, as well as to optimize communication strategies and to learn from the experience gained during usage, are all capabilities made possible by computational intelligence. The cognitive capability of vehicles is improved using technologies such as machine learning, deep reinforcement learning, bio-inspired optimization algorithms, and fuzzy inference systems in an intelligent network. Even more, the vehicle network with intelligent vehicles is amplified in scalability, speed, and reliability by the integration of 5G technologies as well as emerging 6G technologies, cloud computing, etc. Such networks enable very critical applications, including collision avoidance, cooperative adaptive cruise control, dynamic traffic signal optimization, and emergency vehicle prioritization. Vehicle networks due to dynamic topology have unique challenges in latency, scalability, security, and reliability and hence have a need for robust computational models that address these issues efficiently. With transportation in the focus of intelligent vehicle networks, the future is seen on the basis of the way the field has converged and will continue to converge, which is in seamlessly connected, autonomous, and intelligent ecosystems of vehicles and infrastructure working harmoniously in collaboration to optimise mobility and safety outcomes. There is ongoing research and development in this field, which continues to discover unprecedented opportunities to build high adaptive, efficient, and user-friendly transportation systems, making intelligent vehicle networks a key enabler in a sustainable framework for smart cities and urban mobility.

1.1.2 Role of Computational Intelligence in Connected Mobility

The transformative role of computational intelligence in connecting the mobility landscape is to provide the cognitive...