Intelligent transportation systems, representing core components of smart cities, combine sensing, computing, and wireless communication technologies to provide efficient and convenient mobility. Road intelligent transportation systems (ITS) have been studied in recent decades through simulation platforms that integrate computational models for various use cases, such as traffic lights control and management. However, semantic descriptions, illustrating road ITS as a whole and creating a basis for designing simulation platforms, have not been addressed. In this paper, a semantic model of road intelligent transportation systems is proposed, which describes road ITS components and architecture, serving as a formal basis for road ITS simulation platforms. Building upon the semantic model, an extension of the Industry Foundation Classes schema, facilitating ITS simulation platforms based on building information modeling, is discussed. The paper concludes with a summary, a discussion, and an outlook on potential formalization efforts.
traffic simulation and sensor networks components, and a guideline on how to obtain ITS
modeling language for different simulation scenarios, e.g. traffic lights management.
Despite the extensive research on simulation platforms for traffic-related applications and urban traffic management, a formal description of road intelligent transportation systems, providing a basis for simulation platform designs, has received little attention. Building information modeling (BIM), may utilize open standardized data formats, i.e. the Industry Foundation Classes (IFC) standard, to formally describe information and to facilitate information exchange. However, the current IFC schema contains limited entities to be used for defining road ITS components. This paper presents a semantic modeling approach for describing road ITS models. First, background information relevant to road ITS semantic modeling is given. Next, a semantic model of road intelligent transportation systems is presented, followed by an ITS case study, devised to validate the semantic model. The paper concludes with a summary and a discussion on a potential IFC schema extension to be employed for road ITS simulation platforms.
For defining the semantic model of road intelligent transportation systems, properties of all physical and computational components of vehicular environments are to be formally described. In this section, sources that provide knowledge on the physical and computational components relevant to road ITS are analyzed with respect to semantic modeling. The knowledge sources are categorized into (i) network architectures, (ii) road ITS applications, (iii) intelligent road infrastructure, and (iv) communication networks and are briefly explained in the following subsections.
A vehicular ad-hoc network (VANET), representing the fundamental road ITS architecture, is
a network paradigm based on peer-to-peer communications, i.e. V2X (vehicle-to-anything)
communications. As a type of mobile ad-hoc networks, VANETs are self-forming networks
with autonomous and intermittent connections, leading to frequent changes in ITS topology
Merging VANETs with the Internet of Vehicles – the Internet of Things in vehicular
environments – initiated the idea of employing underutilized resources of vehicles, i.e.
onboard units, to decentralize computing processes and provide location-based services.
Network nodes that were solely data consumers have become data producers as well;
therefore, the cloud computing concept together with the VANET paradigm has created the
vehicular cloud concept.
Advances in the automotive industry have led to the design of vehicles with various
computational on-board capabilities, such as powerful computing units, several types of
sensors, and different communication devices. Therefore, vehicles can absorb information
from the environment, perform computational processes, and operate accordingly. As a result,
all network nodes, including vehicles, are able to perform decentralized data processing, i.e.
edge computing, to handle time-sensitive operations more efficiently, and to record data with
local relevance temporarily
Road ITS applications are designed to advance safe and convenient mobility scenarios, i.e.
providing safety to ITS users and goods, decreasing adverse environmental impacts, and
increasing efficient commutes with respect to energy consumption and travel time
Road ITS applications use the data recorded and processed by network nodes. Therefore, road ITS applications are highly dependent on functional elements and on-board resources, i.e. ITS stations that together build the ITS architecture. Intelligent roads comprise ITS stations equipped with various types of devices that have sensing/actuating, computing and communication capabilities. ITS stations are of mobile or fixed type and, according to ETSI EN 302 665, can be further categorized into:
Personal ITS stations, i.e. smart devices, e.g. laptops, tablets, and smart phones (mobile). Roadside ITS stations, which comprise, e.g., traffic shields, cameras, and poles (fixed).
(mobile). Depending on the category, an ITS station may include different functional components and devices. It is assumed that all ITS stations comprise four main on-board units: Sensing unit, computing unit, communication unit, and power unit. The sensing unit contains sensing technologies relevant to traffic detection systems, e.g. micro-electro-mechanical systems (MEMS), inductive loops, radio-frequency identification (RFID), light detection and ranging (LIDAR) sensors, and automatic license plate recognition (ALPR) cameras, or sensor systems to detect environmental changes, such as pollution and temperature sensors. The computing unit comprises main processing and storage devices, while communication devices, e.g. radio transceivers, beacons, and Wi-Fi routers are parts of the communication unit. The power unit includes different power supply means, such as photovoltaic (solar) panels and piezoelectric transducers that may co-exist with electrical grids to deliver energy to ITS stations. In addition, it is worth noting that roadside ITS stations may comprise actuators and control devices to change traffic signals and to control traffic flow on specific routes and road structures (e.g. bridges). Also, most roadside ITS stations provide gateways to the Internet and are therefore considered access points for vehicular clouds. In this paper, regardless of the underlying technology, intelligent road infrastructure is mapped into the semantic model primarily in terms of the abovementioned categories with sensing, computing, communication, and power units.
Advances in wireless network technologies have enabled integrating on-board wireless
communication capabilities into vehicles. Therefore, vehicles can cooperatively connect to
network nodes via V2X communications, i.e. cooperative-ITS (C-ITS) communications
The external network architecture includes an ITS domain and a generic domain, as shown in Figure 2. In the ITS domain, the ITS ad-hoc network represents wireless C-ITS communications between vehicle, personal, and roadside ITS stations. The ITS access network interlinks roadside and central ITS stations and provides communication between vehicle ITS stations through roadside ITS stations. In the generic domain, public access networks grant general data services and applications to public users, whereas private access networks provide secure access only to authorized groups of users.
The internal network in an ITS station interlocks functional networking components of the ITS station. A reference architecture for ITS station internal networks based on layered communication protocols is shown in Figure 3 (ETSI EN 302 665, ETSI TR 101 607). The reference architecture comprises six layers, which characterize different functionalities within ITS stations, and interfaces between layers. The application layer contains standards for road safety, traffic efficiency, e.g. road hazard signaling and collision risk warning, and other applications. The facilities layer includes maintenance of applications, the decentralized environmental notification-based service, communication channels selection, and session supports. The networking and transport layer comprises one or more networking protocols (e.g. GeoNetworking), one or more transport protocols (i.e. dedicated ITSC transport protocols), and a layer management entity. The access layer contains station-internal and station-external interfaces, and specifications for V2X communications in 5.9 GHz frequency band (ITS-G5). Finally, having interfaces with all other layers, the management layer and the security layer contain functionalities that grant ITS station cross-layer and regulatory management as well as intrusion and authorization management, respectively.
Upon analyzing the knowledge sources of road intelligent transportation systems elucidated in the previous section, Figure 4 presents an extract of the proposed semantic model developed from the aforementioned knowledge sources. The semantic model is shown in terms of a class diagram, in which, for the sake of clarity, attributes, methods, and multiplicities of associations are omitted. In the following paragraphs, main elements of the semantic model are described. The DigitalRoad class is composed of RoadStructure and ITSStation classes. The RoadStructure class represents the physical road structure including, e.g. pavement and resting areas. The abstract class ITSStation represents the core elements of intelligent road infrastructure. ITS stations, as mentioned previously, are categorized into two groups, depicted by the abstract subclasses Fixed and Mobile. Fixed ITS stations are of two types, Central and Roadside, and mobile ITS stations are categorized into Vehicle and Personal classes. The SensingUnit, ComputingUnit, CommunicationUnit, and PowerUnit depict ITS station on-board units and are shown in Figure 4 exemplarily for the Roadside class. ITS stations may have one or more control devices, such as Actuator devices.
The abstract class ITSCommunication represents communication networks in the ITS domain and is connected to the abstract class ITSStation with a composition relationship, since communication in the ITS domain is fully dependent on ITS stations. The abstract class ITSCommunication is a superclass of InternalNetwork and ExternalNetwork, which indicate communication within ITS stations and communication between ITS stations, respectively. The InternalNetwork class is defined based on the ITS station reference architecture introduced earlier. Therefore, the InternalNetwork class implements the Reference Architecture interface. Moreover, in internal networks of ITS stations, linked on-board units are recognized by the ProprietaryNetwork class. It is worth to note that proprietary networks comprise all on-board resources, e.g. sensors, beacons and transceivers, mechanical and/or electrical actuators, and several other devices, which are connected to the ITS station. The ExternalNetwork class comprises C-ITS and infrastructure-to-infrastructure (I2I) communications, shown by the V2X and I2I classes, respectively. I2I communication can be either wired or wireless, while V2X communication is solely wireless. The abstract class Wireless is a superclass of LongRange and ShortRange subclasses, which represent different wireless communication standards. An illustrative example of standards in each subclass is given by the Cellular and ITS-G5 classes, respectively.
For validating the semantic model and, specifically, for better understanding the relationship between ITSCommunication and ITSStation classes, a scenario of vehicular cloud formation is considered as a case study. In the case study, communications in the ITS domain are employed for road safety and traffic management applications. As shown in Figure 5, it is assumed that an accident, rear-end collision, occurs in a section of a road, and ITS station external networks form a vehicular cloud to broadcast safety-related messages to other nodes. As can be seen from the scenario in Figure 5, vehicle ITS station V1 broadcasts collision risk signals to ITS stations with potential interests, e.g. vehicles approaching and roadside units in the vicinity. V1 sends a warning message (M1) to the roadside ITS station RSU2 using ITSG5 channels allocated for road safety messaging services. Meanwhile, V1 utilizes Bluetooth wireless communication to inform vehicle ITS station V2 with a collision risk message (M2). Using traffic management applications, V1 sends a message (M3) to roadside ITS station RSU1, requesting the traffic status of an alternative route. RSU2, using 4G cellular communications, broadcasts congestion warnings to inform vehicles in farther distances. In response, V2 asks for media footage of the accident via ITS-G5 (M4). In addition, the roadside ITS stations RSU1 and RSU2 communicate through 4G cellular communications to perform traffic safety-related actions, such as changing traffic lights (M5).
Figure 6 describes the semantic illustration of the scenario depicted in Figure 5 in terms of an object diagram as an instance of the proposed semantic model. The object diagram comprises four ITS station objects, messages shared among ITS stations (M1…M5), and the wireless communication standards employed for each communication.
As has been demonstrated in this section, the proposed semantic model is able to describe cloud formation scenarios with different numbers of ITS stations connected, various communication types involved, and different on-board units engaged. Moreover, data processing and computational models for various applications may be described using the semantic model, representing a means supporting formal descriptions of road intelligent transportation systems. In future work, the semantic model may be mapped into a standardized metamodel to describe road intelligent transportation systems on a standardized basis, preferably employing building information modeling as an increasingly used method in infrastructure modeling. When using the IFC standard as a formal basis, some entities defined in the IFC schema, such as IfcCommunicationsApplicance, may be used to describe road ITS components. However, due to a lack of entities to describe ITS-related infrastructure and components, the current IFC schema cannot be employed for fully mapping the proposed semantic model. Instead, an IFC extension is to be developed to consider all aspects covered by the semantic model of road intelligent transportation systems.
To achieve a formal description of road intelligent transportation systems, a semantic model of road intelligent transportation systems has been developed. Knowledge sources for semantic modeling of road intelligent transportation systems have been analyzed. Associations and relationships between elements of intelligent road infrastructure and vehicular cloud infrastructure have been defined on a meta level using object-oriented modeling, and a scenario of vehicular cloud formation and data-sharing processes has been depicted for validating the proposed semantic model. The model is applicable for various road ITS use cases, and it can be used as a basis for designing ITS simulation platforms. In future work, extending the IFC schema is envisaged to facilitate standardized description of road intelligent transportation systems in terms of building information models. This research is partially supported by the European Union through the European Social Funds (ESF) and by the Thuringian Ministry for Economic Affairs, Science and Digital Society (TMWWDG) under grant 2017 FGR 0068. The financial support is gratefully appreciated. Any opinions, findings, conclusions, or recommendations expressed in this paper are those of the authors and do not necessarily reflect the views of the aforementioned institutions.