Intelligent Cyber-Physical Systems for Autonomous Transportation

As the industrial revolution progresses, cars, airplanes, and railways became the prominent mode of transportation. With the passage of time, all these transportation systems were automated and mechanized, but they came up with some limitations and challenges to address. This calls for a comprehensive understanding of the transportation systems from several points of views. It requires considering these systems technology-wise, their effect on the economy, and its consequences on the socio-environment. Over the years, there have been significant enhancements in transport systems. Intelligent Transportation System (ITS) is an uprising practice of Information and Communication Technology (ICT) which is further based on Inter-Vehicular Communication (IVC). ITS comprises applications that is used to advance the traffic management, reduce congestion, and lessen the road accidents. ITS plays a very crucial role in the economic development of both the industrialized countries as well as the developing nations. Proper and apposite implementation of ITS can effectively reduce the fuel consumption enabling people to travel in a secure and inexpensive mode. This chapter discusses in detail about the older transport system and the intelligent transport system.

Transportation is an integral and fundamental part of human beings’ lives. On Earth, we need transportation in the form of cars, buses, trains, etc. We need aircraft in the air and ships at sea for long-distance transportation. We need space shuttles in space to travel beyond the air. The main desire of human beings is to complete tasks with less energy, effort, time, and more security. The whole paradigm of humanity’s lifestyle can be shifted by autonomous transport (AT), which is already deployed in different technologically advanced countries. The Autonomous Transport System (ATS) is more secure and reliable than the current system of conventional transportation. With the aid of machine learning (ML), artificial intelligence (AI), and blockchain technologies, ultra-fast processing computers can make autonomous vehicles smarter, safer, and more secure than ever before. Connecting vehicles can communicate with the infrastructure to alert the driver about events such as when a train is coming, when a driver cannot see or hear the approaching train, etc. ATS can have a tremendous effect on all we do. However, there are certain challenges involved with every technology, and once we overcome these problems, these ATS can make life simpler, smarter, and safer. Furthermore, we discuss the challenges and future directions for the ATS.

The Autonomous transportation systems will become the mainstream in the future; it is particularly important to design effectively automatic control modules. This calls for the algorithms with highly adaptive ability of perception and computation. By actively interacting with the surroundings, optimal control decision strategies can be automatically calculated. Although research efforts have been devoted to this domain for many years, the general mathematical optimization methods can only find the approximated solutions inside systems, which makes the performance of the transportation systems difficult to achieve perfectly optimal. In this context, introduction of intelligent technologies represented by artificial intelligence has been regarded as a promising perspective. Artificial intelligence (AI), a branch of computer science, attempts to understand the essence of intelligence and simulates a novel intelligent machine that can respond in the ways similar to human intelligence. Since the birth of AI, its theory and technology have become increasingly mature, expanding applications in many fields such as robotics, language recognition, image recognition, natural language processing, and expert systems. It can be imagined that the scientific and technological products brought by artificial intelligence in the future will be the “container” of human intelligence. Generalized into transportation systems, the AI abstracts the complex system processes as black boxes and then uses the idea of statistical learning to model the complex system processes. By discovering potential and unobservable patterns, the AI provides more opportunities to solve the uncertainty problems hidden in the transportation systems. Predictably, the AI technology will be the key breakthrough in the evolution from conventional transportation systems to autonomous transportation systems. Therefore, this chapter is organized via three aspects of contents: (1) Overview of AI; (2) Need and Evolution of AI; and (3) AI for Transportation Systems. In this chapter, the wide AI conception is described by dividing it into three branches: cognitive AI, machine learning AI, and deep learning AI. Further, the need and evolution of AI are surveyed via three parts. The first part describes origin and development of AI, the second part describes common approaches and technologies about AI, and the third part describes successive cross-field applications of AI. Finally, application of AI for transportation systems is introduced via three parts: motivation, application status, and challenges.

The decision-making power of machine working on human behavior through perception, learning, and reasoning is called Artificial Intelligence (AI). AI tries to break the barrier between human and robots as an expert machine. Artificial Intelligence is useful in order to save time and increase the productivity of work. AI automates routine tasks which were previously performed by human beings, consequently, the cost of hiring human resource is decreased. Furthermore, with the help of AI, such machines can be developed which will be much faster and smarter than human beings. The idea of modern AI was first presented by John McCarthy in Dartmouth conference in 1956. The first golden years of AI from 1956 to 1974 were at that time when AI was considered quite interesting and a hot topic for researchers and it was believed to be good investment, time and money. However, at the end of 1974 the first winter of AI was started, it was considered that AI is not as such of great progress rather wastage of time and money. Afterwards, another rebirth of AI from 1980 to 1987 took place with the knowledge about the expert systems. In 1997, a supercomputer named as Deep blue was created on the basis of AI. After that DARPA Grand Challenge in March 2004, it was about an autonomous vehicle drive through uncharted terrain by itself for hundred miles that vehicle was also working on AI. Hence, transportation system started using Artificial Intelligence. With AI being part of the system, congestion, accidents can be controlled, predict future moves (e.g., RATP dev AI engine works on it). In addition, AI is used for scheduling and routing, applying higher and lower traffic patterns. In ITS specifically in VANET (Vehicular Ad hoc Networks) there are two modes of communications V2V (vehicle to vehicle) and V2I (vehicle to infrastructure). AI is also used in Swiss Federal Railways and Tesla self-driving cars. According to UITP’s survey it is stated that 62% of public transport players are involved in Artificial Intelligence projects and 86% are engaged in partnership to develop and adopt AI. In this chapter, authors discuss Artificial Intelligence, its needs, and evolution in transportation system. In addition, the challenges and future directions are also described in detail.

Artificial intelligence (AI) is a concept in which entities and systems have the ability to learning and decision-making by imitating biological processes. In this chapter, we first introduce the evolution of AI to give the reader a good grasp of artificial intelligence (AI) and then introduce the existing machine learning (ML) with three typical classifications: unsupervised learning, supervised learning, and reinforcement learning. Supervised learning makes decision based on the output labels provided in training. Unsupervised learning works based on pattern discovery without having the pre-knowledge of output labels. The third machine learning paradigm is reinforcement learning (RL), which takes sequential actions rooted in Markov Decision Process (MDP) with a rewarding or penalizing criterion. Some other kinds of ML algorithms such as federated learning and transfer learning are also introduced in the first subsection. The future transportation network aims to develop a highly dynamic and intelligent system, which enables the networks to change the environment to satisfy various requirements and service types. Cellular-V2X, vehicular edge network, and unmanned aerial vehicle (UAV) are recently attacking network architecture to enable the future transportation. In the second section, we introduce how to integrate AI into cellular-V2X, vehicular edge network, and UAV. Leveraging AI into transportation helps the sector increase passenger safety, reduce traffic congestion and accidents, lessen carbon emissions, and also minimize the overall financial expenses. Finally, we review some existing research that uses AI to enable autonomous driving, traffic control and prediction, and path planning.

Transportation science is a research field with many categories, including information science, control science, Earth science, management science, etc. How to integrate different areas to make all the components work efficiently acts as a challenging task. Characterized by the strong modeling and computation ability, AI technologies have been applied to many areas in contemporary society. To this end, the deployment of artificial intelligence (AI) technology into transportation systems leads to a typical form of intelligent transportation systems: AI-Deployment transportation systems (AID-TS). The AID-TS effectively apply advanced AI technology into transportation, service control, and vehicle manufacturing, in order to strengthen the connections among vehicles, roads, and users. To this end, real-time autonomous transportation systems with considerable efficiency and reliable safety can be formulated. Through the harmonious and close cooperation among people, vehicles, and roads, the AID-TS are able to improve transportation efficiency, alleviate traffic congestion, improve capacity of road networks, and reduce energy consumption. Nowadays, countries represented by Japan and the United States have widely studied application of the AID-TS. At the same time, China is also actively planning its research on AID-TS. Generally speaking, AID-TS will become a wide and meaningful application in the future smart city around the world. From the perspective of system composition, the AID-TS are a kind of complicated and comprehensive systems that can be mainly divided into several subsystems: transportation information service systems, transportation management systems, public transportation systems, and vehicular control systems. Therefore, this chapter is organized via three aspects of contents: 1) review for AID-TS; 2) architecture for AID-TS; and 3) business scenarios for AID-TS. This chapter is responsible for introducing basic conception of AID-TS and is composed of three parts: overview, prevalence, and development status. In this chapter, architecture for AID-TS as well as key technologies is described. From the bottom to the top, the whole architecture can be categorized into three layers: sensing layer, networking layer, and application layer. This chapter describes four typical business scenarios for AID-TS in four parts, containing autonomous transportation management, vehicular control, public transportation scheduling, and transportation information service.

The revolutionary research and experiments in the field of computing and communicating technologies have resulted in a dramatic impact on the applications with societal and economic benefit. With the evolution of the Internet, it is now possible to connect every object or thing in the physical world. These things can communicate and also perform computations. It is now possible for humans to communicate with the physical things around us. This leads us to explore a whole new technology called Cyber Physical System (CPS). CPS combines computation, communication and control technologies in order to integrate the existing networked systems and embedded systems. It has modules to perform accurate data acquisition. These modules are basically distributed devices. Further the acquired data is sent to a layer of information processing as per the service requirements. There are enormous applications of CPS namely: digital medical devices, autonomous vehicles, robotic systems, intelligent highways, aerospace systems, industry automation, building and environment control and physical process control. Among the listed applications, autonomous transportation is evolving through the ongoing research trends. In autonomous transportation systems, there is a high requirement for reliable communication between the communicating entities, accurate data acquisition and processing and high computing capabilities. Thus CPS is a novel engineering system that can suit the requirements of autonomous transportation. In this chapter we discuss the evolution of CPS and its role in future autonomous transportation. We explore the research challenges in the CPS based on autonomous transportation.

This chapter presents the historical evolution and a prospective vision of the Cyber-Physical Systems (CPS), with a specific focus on Intelligent Transportation Systems (ITS). Transportation Cyber-Physical Systems (TCPS) enhance transportation systems’ efficiency and reliability by enhancing feedback-based interactions between the cyber and physical systems. In addition to an introduction to CPS and TCPS, this chapter illustrates the architecture and advanced technologies to achieve traffic intelligence. Next, we discuss the technical challenges and highlights feasible solutions. Finally, we demonstrate the significant roles of CPS in ITS and conclude the chapter.

The estimates state that more than 75% of Earth’s population will be living in urban areas by 2050. As the majority of population belonging to rural areas have started migrating to urban areas, this has led towards the issues of higher pressure on the existing resources and the infrastructure. These trends could be easily noticed in China, India, Africa, and Australia. In vehicles, vehicle-to-vehicle and vehicle-to-infrastructure applications tend to employ DSRC and Vehicular Ad Hoc Networks (VANETs) to provide rich, real-time fleet data to improve safety, efficiency, reliability, comfort, and convenience that would be infeasible to collect within the confines of a single vehicle. The work focuses upon a specific application of Intelligent Traffic Light System (ITLS) while keeping into consideration the need of traffic light clearance for emergency vehicles at the earliest possible. Future work of the paper includes evaluating the efficiency of proposed scheme and working on an offline–online sync-based information dissemination mechanism in a smart city.

Ocean exploration has received much attention lately and became very important research area since last one decade for pollution detection, target tracking, navigation and ocean monitoring for military surveillance purposes. For underwater communication network of sensors needs to be established to collect information. Underwater networks are different than territorial networks because of dynamic underwater environments and both have different sort of challenges. UWSNs have challenges like battery replacement, node localization, end to end delay, more data loss and high signal-to-noise ratio. Network topologies are considered dynamic in UWSNs because sensors are not stationary and moving with the water current. After deployment of sensor nodes, routing protocols plays vital role for secure and efficient communication. In UWSNs network lifetime and power consumption are considered major challenges in the way of underwater communication. Since last decade many protocols are presented to coupe up with these challenges efficiently in dynamic scenarios. In this chapter we have presented a protocol which predicts fast moving nodes and neighbor nodes to exclude them from clusters. In this way the skipped nodes became idle and conserve their energy. In next round of cluster formation residual energy level of idle nodes is found higher than used nodes so they will form cluster. Neighbor nodes cover overlapping area and used to collect a degree of matching information which later eliminated at aggregation level. Our proposed algorithm finds speed of moving nodes to differentiate the suitable and unsuitable nodes for the cluster. In addition we have also differentiated long range and short range communication for acoustic and optical mediums, which also helps to increase network life span. We have used advantages of both communication mediums for short range and long range communication where are they performed well.

Intelligent transportation is related to the developments in various types of technology. In this chapter we want to discuss the use of deep learning model to predict potential vibrations of high-speed trains. In our research we have tested developed deep learning model to predict potential vibrations. We have tested various time steps and potential error margins. Results of our research show that our system is able to predict with accuracy of above 95% with precise results in a series of values forward.

The objectives are to explore the advanced complex data analytics of autonomous transportation in smart cities, improve the industrial production efficiency and product quality, and optimize the industrial production processes. Aiming at the deployment problem of software-defined network (SDN) in distributed cloud data center, the study proposes a spectrum-based SDN deployment algorithm. An industrial complex data model is built. In addition, a complex event processing model is implemented based on the emerging stream data flow execution engine and the emerging Flink stream processing framework. The Flink framework not only performs real-time analysis of fault tolerance but also greatly simplifies the data processing process. Tests of multiple sets of industrial data from iron mine, chemical, and metallurgical plants have proved the operational advantages of the complex event model for the analysis of industrial complex data. Combining the advantages of online filtering, this study proposes a Generalized Robust Least Square (GRLS) data coordination algorithm. The results confirm that the GRLS has better robustness in correcting significant errors. The spectrum-based SDN deployment algorithm solves the problem of SDN domain partition and controller deployment. This proves that complex event processing technology can extract more valuable information from massive and complex data, thereby ensuring the stability of autonomous transportation.

The primary objective of an autonomous vehicle (AV) is to provide a safe and driverless vehicle experience on the road. To accomplish collision free journey, an AV collects data from external entities such as camera, Lidar, radar, etc. and passes this data to a safety algorithm. This algorithm detects and monitors the steering, acceleration, and braking behavior of the vehicle. This can be a very dangerous situation if these internal components which control the behavior of AV fail to accept the instructions. Moreover, if there is a collision on the road, the AV must communicate and transfer data (position, direction, speed, alerts, etc.) with other AVs. There is a need to detect any type of failure in the AV in advance to ensure safety of all the users of the road. This chapter proposes a collision avoidance system for AVs. The proposed system introduces the concept of Meta Sensor in the AVs with fast communication technology Li-Fi to forward the message to leading vehicle (LV) to avoid collision. The aim is to detect the internal sensors behavior in advance with the use of Meta Sensor to avoid collisions. As soon, the failure of any internal sensor is detected, other AVs are immediately informed through exchange of messages and Li-Fi Technology. Emulation using NetLogo shows that in case of a sensor failure, an accident situation is avoided in minimum amount of time.

Cyber Physical System (CPS) is considered as an emerging mechanism where physical devices are managed and processed through automated computations. The Smart vehicular technology is considered as one of the emergent applications in CPS where the involvement of malevolent devices leads to reduce fully adoption risk by various organizations. Though researchers have proposed various cryptographic techniques in order to ensure a secure information transmission among vehicles, however, the existing technique may further leads to several storage, transparent, and cost efficient issues. Though researchers have proposed various blockchain schemes, however, there exist some severe issues that need to address. Further, number of miners elected to validate the block or transmitted information may further lead to increase the computation time and complexity in the network. The aim of this chapter is to propose a secure information transmission among vehicular devices by blockchain based technique. The proposed system generates the contracts with the gas used to encrypt files. The proposed phenomenon is validated over various security measures against conventional approach.

The increasing number of vehicles has brought great pressure to urban traffic. The development of intelligent transportation provides new ideas for alleviating traffic problems. If drivers can get the road condition of certain road sections, it will greatly improve the driving efficiency and optimize the whole traffic network. Road condition recognition often needs a large amount of road condition data collected by multiple vehicles and aggregated by crowdsensing. However, the data collected by vehicles may carry privacy information, privacy leakage problem will affect the incentive of car owners to complete the crowdsensing tasks. In order to protect the privacy and security of the collected data, the aggregation of raw data could be replaced by the aggregation of road condition recognition models. On this basis, we propose a new secure crowdsensing algorithm based on federated learning and blockchain. Specifically, the task publisher deploys the crowdsensing tasks on the blockchain, and the leader of each team in the crowds decides whether to accept the task. When a team leader accepts the task, the team members train the road state recognition model, and the team leader aggregates the models provided by the team members through federated learning. In addition, each member adds differential privacy noise to its own model to further protect privacy. The team rewards each member based on the accuracy of the traffic recognition model provided by the member. Experimental results show that the proposed strategy performs well in road condition recognition accuracy for intelligent transportation.