Advances in smart roads for future smart cities

Abstract

Various countries throughout the world have started their efforts in designing and implementing smart cities. China alone has over 300 smart city projects, with strong participation by industries and government offices. India too have allocated trillions in budget to build over 100 smart cities. An essential part of a smart city is transport. In this paper, we will discuss the current state, developments, and some of the emerging advances in transportation technologies and how these advances in smart roads will prepare the society towards the realization of future smart cities.

1. Introduction

Technologies have continued to impact society over the years, improving our standard of living and quality of lives. Advances in telecommunications, Internet of Things, cloud and edge computing, scalable storage, and data analytics have made fast computing, data-empowered insights, connected mobility and anytime communications possible. With added capability from a fusion of emerging technologies, many countries have now introduced national projects related to smart cities in order to transform lives, enhance business operations and market competitiveness. There are many definitions of smart cities today, including those from standardization bodies. For example, the British Standards Institute [1] defines it as

An effective integration of physical, digital, and human systems in the built environment to deliver a sustainable, prosperous, and inclusive future for its citizens.

The Chinese national smart cities working group has defined it as

Smart Cities bring a new concept and model, which applies the new generation of information technologies, such as Internet of Things, cloud computing, big data, and space-geographical information integration, to facilitate the planning, construction, management, and smart services of cities.

In fact, the ITU-T focus group has analysed over 100 definitions and ultimately come up with its own definition [2], which is

A smart sustainable city is an innovative city that uses information and communication technologies (ICT) and other means to improve quality of life, efficiency of urban operations, and services, and competitiveness, while ensuring that it meets the needs of present and future generations with respect to economic, social, and environmental aspects.

The key technologies behind smart cities are connectivity, cloud computing, data analytics, sensors, Internet of Things and artificial intelligence. Smart cities cover a wide range of applications and used cases. The three commonly used cases highlighted by many countries are (i) transport, (ii) health, and (iii) living. In this paper, we shall focus solely on transport for smart cities. In particular, we shall examine the advances over the past decades on smart roads globally.

Transportation forms the arteries for modern society and economy. The transportation of goods and people has enabled business success and created new cities. While transportation is commonly viewed as a classical civil and structural engineering problem, it is increasingly becoming digitally enabled with info-communication technologies. Current traffic problems facing our society today include: (i) traffic jams, (ii) accidents, (iii) pollution, (iv) fuel cost, (v) fuel scarcity, (vi) high insurance costs, and (vii) others. The increase in population in cities and in the number of cars, bicycles, motorbikes and road users have added to the risk of accidents, traffic congestion, etc. Hence, there are several advances made over the past decades to address some of these problems.

Smart transport includes: (i) smart roads, (ii) smart street lights, (iii) smart cars, and (iv) smart traffic signs. In the quest for zero carbon emissions and overcoming high gasoline prices, electric vehicles (EVs) are commonly sighted on roads these days. Also, with the push from large corporations like GOOGLE and UBER, self-driving autonomous vehicles (AVs) are actively being designed, created and tested. Some companies even go as far as designing and constructing flying cars [3]. All these are further fuelled by companies working on advanced driver-assistance systems (ADAS), vehicle charging solutions and semiconductor companies producing powerful artificial intelligence chipsets.

As shown in figure 1, various transportation-related technologies have evolved over the years since the mid-1990s, covering an interesting wide spectrum of areas addressed by these technologies. Starting from melody roads, to the emergence of vehicle-to-vehicle communications, vehicular ad hoc networks, electrified roads, harvesting of energy from roads, smart road intersections, self-weighing roads, ITS cooperative emergency rescue, methods to capture driving behaviour, smart street lights and wireless digital traffic signs.

Figure 1.

Timeline of technology trends related to transportation for the last two decades. (Online version in colour.)

To gauge the level of research intensity by the community, we have performed Google search on these topics using keywords and the results we obtained are shown in table 1. As shown, AVs and smart cities have generated a lot of attention and research work from the community lately. Intelligent traffic signs and traffic intersections too are popular research topics. Traffic road safety alone tops the list with over a million publications. Hence, traffic safety is still the core for many researchers in the transportation field. While many industry leaders have predicted the need for self-driving vehicles in our future roads and several trials have been made in California, Arizona and other states, traffic road safety is still of paramount importance.

Table 1.

Hits using keywords on GOOGLE SCHOLAR.

keywords	hits
melody roads	9
road energy harvesting	55
vehicle mobile sensors	58
traffic accident emergency rescue	73
smart intersections	143
smart street lights	236
smart highways	1350
flying cars	2570
smart roads	2700
intelligent roads	3350
smart cars	8200
self-driving cars	16 100
mobile sensing	16 900
VANETs	34 500
driving behaviour	71 300
smart cities	82 700
smart intersections	92 400
intelligent traffic intersections	120 000
autonomous vehicles	128 000
intelligent traffic signs	439 000
traffic road safety	1 890 000

Instead of covering EVs and smart cities, in this paper, we shall focus on smart transport. Specifically, we shall discuss how advances made in smart roads have evolved over the years and how they will help in realizing smart transport for future smart cities (figure 2 and table 2).

Figure 2.

Future smart transport is about intelligent vehicles, signs and roads. (Online version in colour.)

Table 2.

Summary of 10 recent advances in smart roads.

number	advances	sector
1	roads that harvest and store energy	energy
2	roads that sing: musical road	safety
3	roads that weigh your car	logistics
4	roads that charge: electrified road	energy
5	roads with wireless digital traffic signs	safety
6	roads that detect traffic violations	law enforcement
7	roads that ‘talk’ (V2X)	safety, logistics, law enforcement
8	roads with smart intersections	safety
9	roads with fast emergency rescue	safety
10	roads with smart street lights	citizen services

2. Advances in smart roads

In [4], the future of smart transport is about intelligent signs and roads. In fact, we are entering the era where the information superhighway (connectivity, Internet and data grid) meets the transportation highway. Roads are no longer viewed as merely a physical entity or solid ground. They will be ‘empowered’ with info-communications, intelligence and sensing capability that were never possible decades ago. Work done on intelligent transportation system (ITS) by standardization bodies include ISO TC 204, IEEE 802.11WAVE, car-to-car consortium, etc., while most of the early standardization work was focused on air interfaces, there is an increasing need to look at architectures, systems and applications.

In the following sections, we will discuss ten advances that have occurred over the years for smart roads. Although these advances address a specific problem in transport, it is useful to collectively understand these advances and evaluate how they will influence the future construction of smart roads.

(a). Roads that harvest energy

Roads were meant to provide transportation links from one place to another. Without roads, vehicles will have to ride over bumpy surfaces or made long routes instead of direct routes due to obstructions. However, smart roads of the future will be able to do more than just paving the way to destinations. Ideas of using roads to harvest energy have also emerged recently [5,6].

There are several methods to harvest energy from roads. Some use sunlight (hence the term solar roads [7]) while others use mechanical vibrations produced by vehicles as they transverse the road to generate electrical energy. Solar energy captured on roads can be used to power street lights, signage and traffic signals. The energy harvested can also be stored or fed to the electric power grid. This is the point where transportation power grid meets the electricity generation grid, changing the future of the power supply paradigm. For solar roads, photovoltaic modules are placed directly on top of road surfaces to capture sunlight, as shown in figure 3. The energy generated can also be used to light up street dividers during the night and for melting snow and ice during winter. Tables 3 and 4 show the countries that have already experimented and implemented solar roads and ‘piezoelectric roads’.

Figure 3.

Solar panel road tiles (source: Solar Roadways Inc). (Online version in colour.)

Table 3.

Deployment of solar roads in China, France and USA.

country	solar roads
China	Jinan, Shadong province—2 km road with solar modules
France	Wattway solar road to power toll gates and payment machines
USA	Solar Roadways is an American startup working on generating electricity from roads and replacing concrete and asphalt roads with solar panels [8]. Each tile is a 44 Watts solar panel coated with a tempered glass, capable of withstanding the weight of a semi-truck. LED lights on the panel can serve as signage or lane markers. The panels are modular and can be easily replaced during maintenance.

Table 4.

Deployment of piezoelectric roads throughout the world.

country	piezoelectric roads
USA	Invested $2.3 M on projects related to piezoelectric roads; efforts are led by California Energy Commission and Oregon DOT In Washington DC, 240 m2 pavement powdered by footprints
UK	Lancaster University research programme on generating power from passing road traffic, with the goal of generating 1–2 MW/km at traffic volume of 3000 cars/h.Bird's Street, Central London, piezoelectric roads installed to power street lights
Japan	East Japan Railway Company—2008–2009 deployment at Tokyo station Marunouchi North Exit. Production of electricity reached 10 K Watt-second per day. The electricity generated is used to power all electronic displays at the station.
Italy	Piezoelectric floors depleted for Venice-to-Trieste Autostrada.

Piezoelectric roads [9,10] use piezoelectric devices to generate electrical energy. Piezoelectric crystals are placed about 5 cm below the asphalt surface, and these crystals became slightly deformed when vehicles travel across the road. The crystal deformation produces electrical current, as shown in figure 4. Hence, mechanical energy from cars is now converted into electrical energy. Piezoelectric devices have been deployed by the East Japan Railway Company (under subway station gates) and by Innowattech Ltd. (under roads in Israel). According to Innowattech, if such devices are installed along a kilometre stretch of road, an average of 400 KW of power can be harvested, which is enough to power 162 Western-U.S. homes [5,6]. Traffic flow rate (vehicles/h) affects power density and piezoelectric technology performs best in areas with high traffic flow rates.

Figure 4.

Characteristics of a piezoelectric device. (Online version in colour.)

A third form of energy harvesting is through the use of dynamic speed bumps to convert kinetic energy to electric energy from passing vehicles. This was proposed as part of EU FP7 POWERAMP project [11].

(b). Roads that produce music—musical road

Musical roads are roads that can produce music or tunes when cars drive passed them. It is also known as road-as-an-musical-instrument. Countries like Japan, USA, Denmark, Netherlands, Taiwan and South Korea have constructed such roads. Engineers from Hokkaido Industrial Research Institute in Japan have developed a musical road surface, which use cars as tuning forks to create music. The key concept is the use of grooves or rumble strips, which are spaced at specific intervals on the road surface. Hence, depending on inter-spacing of grooves, cars moving over them will generate a series of high or low notes. Designers can then create music based on the variation of these notes. It was found that the optimal speed to produce the soundtrack is 28 mph. Driving too quickly over it is equivalent to playing fast forward.

The purpose of musical roads is not merely for entertainment but also for hazard warning, road safety and helping drivers to keep to the speed limit. As shown in table 5, musical roads have served its safety purpose well in several countries. However, such music can be disturbing to residents living nearby at nights, as their sleep are often interrupted by the music. Hence, musical roads have to be restricted in the night or they must be located far away from residents. It is particularly suitable for highways and long country roads, to serve as a reminder of their driving speed.

Table 5.

Deployment of musical roads around the world.

country	musical roads and remarks
Japan	Japan has over 30 melody roads, located in Hokkaido, Hiroshima, Shizuoka, Oita, Gunman, etc.
Korea	‘Mary had a little lamb’—for warning drivers on Korean highways
Taiwan	Located in Kinmen's ‘Dinglin Road’, cars driving at 50 kph generate a tune of ‘The Olive Tree’.Anti-slip strips are used instead of permanently cutting grooves on road surfaces.
Denmark	In 1995, Danish artists invented the ‘Asphaltophone’, raised pavement markers that produce tones.Purpose is to keep drivers awake and observe the speed limit
Netherlands	In Leeuwarden, road will play out a tune at 40 mph speed limit of the Friesland's regional anthem.
USA	New Mexico—On historic Route 66 between Albuquerque and Tijeras, the song ‘America the Beautiful’ can be heard by drivers at speed limit of 45 mphLancaster, CA—a snippet of ‘William Tell Overture’ plays for drivers at 55 mph.

(c). Roads that automatically weigh your car/truck

Roads are not only used for transporting people but also heavily used for transportation of goods. For example, in the USA alone, billions of tonnage are being shipped yearly, as shown in figure 5a. Road transportation is an essential part for the freight business in many countries. Overloaded trucks pose a potential hazard on the roads, often tumbling at high speeds, sudden brakes or manoeuvering sharp turns and corners, as shown in figure 6a. Hence, trucks often need to be checked on their weights and safety compliance. Over the years, advances in technologies have improved static weighing to weigh-in-motion (WIM) [12] and virtual WIM [13,14].

Figure 5.

(a) Total tonnage shipped in the USA from 2015 to 2045 (source: US Department of Transportation, Bureau of Transportation Statistics and Federal Highway Administration, Freight Analysis Framework, v. 4.1, 2016.) and (b) a truck is being weighed to ensure it complies with the regulations. (Online version in colour.)

Figure 6.

(a) Overloaded trucks are prone to severe traffic accidents, causing fatalities and roadway shutdowns and (b) high speed weight-in-motion system with road sensors in action. (Online version in colour.)

(i). Static weighing

While normal cars do not need to be weighed, other trucks and heavy delivery commercial vehicles will need to be weighed on roads and charged toll fees. Trucks exceeding the legal mass limits increase the risk of traffic accidents and damage to road infrastructures. A weigh station is a checkpoint along a highway that performs this role. In the USA, weight stations are used to collect road usage taxes by commercial freight vehicles at points of entry too. The maximum weight set by the USA federal government is 80 000 pounds. Some trucks have to stop to be weighed while others can be automatically weighed even while passing through. In static weighing, trucks have to stop and weigh over a scale as shown in figure 5b. While effective, at times of heavy traffic, delays build up at the weighing station as trucks queue in line to be inspected prior to being allowed to continue on their road journeys.

(ii). Weight-in-motion

In the 1970s and 1980s, WIM data were used for bridge calibration and assessment, mainly focusing on fatigue and load effects. WIM data were also used for traffic monitoring and statistical analysis on road freight transport. In the 1990s, the first WIM standard (ASTM-1318) [15] was published in North America, and the COST323 [16] action provided European specifications of WIM. In the early 2000s, the accuracy of WIM systems was significantly improved and they were used mostly for overload screening and enforcement. WIM technology is better than static weighing because trucks can automatically be weighed while driving. Hence, unlike static weighting, less delays are present and trucks no longer need to wait in line for inspection. In WIM, multiple sensors are installed in one or more traffic lanes and measurement of axle and vehicle loads are done in real-time while these vehicles are in motion [17].

(iii). High speed weigh-in-motion

HS-WIM technology can weigh vehicles dynamically, with speed up to 80 mph (129 km h⁻¹). Vehicle weighing is done in the traffic lane at existing speed (i.e. 60—90 or 100 km h⁻¹), without the need for slowing down or stopping the vehicle. HS-WIM systems can automatically record and display wheel-load weights, axle weights, gross vehicle weights (GVWs) and other parameters. HS-WIM is well-suited for weight enforcement screening, monitoring of bridge loads, toll roads and for traffic data collection. Minimally invasive in-ground strip scales are placed in grooves of less than 75 mm in the pavement, keeping lane closures to a minimum. These strips are high-performance strain gauges, and are capable of operating over a wide range of environmental conditions. They operate on the principle of measuring the change in resistance, as they are elongated, in relation to the strain of the base (load cell) material.

(iv). Virtual weigh-in-motion technologies

A virtual weigh station (VWS) is an enforcement facility that does not require continuous staffing and is monitored from another location. A roadside WIM server sends data to the cloud, enabling users' remote web access and control of all WIM-related functions. Virtual WIM or V-WIM provides an unattended, unmanned means of automatically collecting data. V-WIM relies on the same HS-WIM technology and is coupled with cameras with optical character recognition, licence plate reader technology and wireless output to collect a variety of traffic data, such as GVW, axle weight, vehicle class, ID and imaging, all of which can be accessed from a remote location. This technology aids engineers and road designers in the study of traffic characteristics, as they are related to traffic flow, pavement design and degradation. V-WIM is capable of determining a commercial vehicle's weight to a degree of accuracy and delivering the vehicle identification and weight data to enforcement personnel in real-time [18]. Both HS-WIM and V-WIM systems are widely deployed in USA (for example in Maryland, Virginia, Idaho), Canada and Taiwan.

(d). Roads that automatically charge your vehicle (aka ‘electrified roads’)

With growing trends to reduce carbon emissions and alleviate fuel shortage, EVs are now more commonly in use. However, EVs need to be recharged if the remaining battery capacity runs low. Often, it may be difficult to find charge stations nearby, especially if one is in the rural areas. Within cities, charge stations are now located at shopping malls and car parks. Also, some school and public buses are now powered totally by electricity. In Korea [19], buses within the campus can be charged automatically and continuously, allowing them to use smaller batteries (a third in size of those in regular electric cars). Other methods charge the vehicle at bus stops and on certain segments of the fixed routine route.

Some cities have suggested the use of specially assigned charging lanes for autonomous electric trucks for the future. Trucks usually cover long distances in order to deliver goods and hence, have to be recharged frequently. In Sweden [20], about 1.2 miles of road near Stockholm has been transformed into an ‘electrified road’. It recharges the batteries of cars and trucks as they traverse on the road. It is part of Sweden's ‘eRoadArlanda’ project, which aims to provide dynamic charging for vehicles instead of using roadside charging posts. This, under the Swedish's proposal, does not rely on the use of rail tracks on the road.

In UK [21], the government is interested in new road technology capable of charging cars as they drive on roads. Specifically, the government is interested in magnetic induction technology. The idea is to have cables buried underneath the road surface so that they would generate electromagnetic fields strong enough to be received by a receiver device in the car, transforming it into electrical power. This principle follows Michael Faraday's laws of electromagnetic induction. Specifically, a green colour lane is allocated for this purpose. Electrified roads have the collective aim for achieving ultra-low emissions from EVs (figure 7).

Figure 7.

Electrified roads that charge vehicles on specifically assigned lanes [19]. (Online version in colour.)

(e). Roads with smart wireless digital traffic signs

Traffic road signs have been in existence and in use for a long time. In fact, the UK and USA are among the first few countries in the world that deploy traffic signs to warn drivers and enforce traffic rules. However, as pointed out in [22], there are issues associated with traffic signs, such as

• (a)poor visibility of traffic signs,
• (b)challenges in placing signs, and
• (c)difficulty in remembering the highway code.

There are over 60 signs in the USA and 170 signs in the UK highway code to study and remember, making it very difficult for senior drivers and those with poor and fading memory. As shown in figure 8, the idea of wireless digital traffic sign is to embed a server into the traffic sign board. The specific sign is then broadcast wirelessly to on-coming traffic. A receiver unit residing in the car will then pick up the wireless sign signal and alert (verbally or on the display) the driver about it. This entirely removes the need for the driver to watch out for signs while driving and he can focus his attention on what is happening ahead of him.

Figure 8.

The programmable wireless digital traffic sign post concept [22]. (Online version in colour.)

There are several advantages of using this new wireless traffic sign: (i) it eliminates the need for the sign to be visible to the human eye, (ii) it removes the load on the driver to watch out for the signs while driving, (iii) it removes the burden for the driver to remember all the traffic signs, (iv) it is not affected by poor weather and lighting conditions, (v) the sign is programmable, meaning changing a sign is as easy as reprogramming it, (vi) there is no need for complex signal processing and image traffic sign recognition that is in used in self-driving cars today, (vii) automatic computation of traffic volume, and (viii) low cost.

Multiple wireless digital traffic signs can exist on roads and hence signal directivity is important as signs are intended for on-coming vehicles. Signs that a vehicle has driven past are no longer important to them. Hence, with signal directivity provided by directional antennas, signs can be directed at multiple on-coming vehicles, across all lanes in the same direction of travel. This new architecture also allows the introduction of several new applications, such as automatic traffic volume measurement and automatic traffic violation detection and citation [22].

(f). Roads with smart traffic violation detection, citation and notification

Traditional methods of detecting traffic violations are done through pre-installed cameras and speed radar detectors with white line markings on the roads [23,24]. These methods are less effective under rain, snow and fog conditions due to poor visibility. Two recent solutions for automatic traffic violation detection have emerged recently: (i) via the use of wireless digital traffic signs [22] and (ii) the used of drones or unmanned aerial vehicle (UAVs) [25].

(i). Traffic monitoring using wireless digital traffic signs

With the presence of wireless traffic sign posts, a new traffic violation detection method can be created [22]. For example, drivers will not be able to deny and ignore the presence of signs to traffic law enforcers, as the received and voice-narrated signs will be recorded as evidence within the car ADAS client system and also on the car black box.

As shown in figure 9, driving at a speed exceeding the stated speed limit will have violated the traffic code, and the wireless digital sign post will alert the traffic citation system, which can then directly send the traffic citation to the driver, via a cellular connection. Alternatively, the violation alert can be sent by the wireless digital traffic sign post directly to the driver, along with sending a copy to the traffic citation system. The wireless traffic sign information received by the diver is recorded as proof of successful notification, and the driver cannot deny not receiving it. This process is automatic and seamless and hence, will enhance the traffic violation enforcement operation efficiency.

Figure 9.

Next generation of smart roads with smart traffic citation system [22,45]. (Online version in colour.)

(ii). Traffic monitoring using drones or UAVs

The second method involves the use of drones or UAVs and this has been recently used by various police forces around the globe, as shown in table 6. The key challenges for detecting traffic violation is the need for evidence. Evidence is needed at the time and location when the violation occurs. Table 6 shows that several countries have adopted the use of drones on the roads. Drones equipped with cameras and wireless communications are effective in video recording of traffic violation incidents, providing identity of cars and drivers and notifying the violation event to the police. In other used cases, drones can also be used to detect on-going crimes, acts of terrorism, drivers texting or talking on the phone while driving, suspicious drivers who are driving under the influence (DUI) and for investigations of post-crashes in road accidents.

Table 6.

Details of the application of drones on the roads for traffic violation detection and enforcement.

country	remarks
France	Bordeaux, France: In 2017, police in France is using camera-equipped drones in the sky to catch drivers who violate traffic rules.
UK	Southwest county: In 2017, UK police is using drones for monitoring traffic accidents, terrorism and crime scenes.
USA	Wentzville, Mo: In 2018, Wentzville police uses drones for traffic monitoring and violations citing for drivers failing to stop at stop signs [25].Maine: Maine state police have been using drones for road accident reconstructions and investigations [26].
China	Jinan, China: In 2018 [27–29], drones are used to detect traffic violation at the capital city of Shandong province.Illegal behaviour caught include: (a) using mobile phones while driving and (b) failure to stop or give way.More than 300 drones are used in 25 provinces in China, with numbers expected to rise to 1000 by 2020 [30].
Israel	Police in Israel [31] has started using drones to record reckless drivers, speed violators and drivers who use their phones while driving.Cell phone usage while driving was the number one cause of car accidents in Israel in 2017 [32].

(g). Roads with V2X and VANETs—cars that talk

Mobile ad hoc networks have penetrated the vehicular space with vehicular ad hoc networks and car-to-car communications capability, or commonly called V2V. Communications from car to any roadside infrastructure is termed as V2I while V2X refers to communications from the vehicle to any other object. V2X is important because it allows cars and objects to communicate and exchange crucial information, be it position, identity, state of physical presence or speed information. With such data, cars can be alerted about potential upcoming traffic hazards, avoiding accidents and enhancing safety for road users.

Currently, V2V can be used to propagate accident alerts to other neighbouring cars. Traffic jams status can also be propagated, advising cars behind to take advance exit and seek alternate routes in order to avoid further congestion [33,34]. It can also be used to detect dangerous drivers (misbehaving drivers or criminals on-the-run) on the roads [35].

For V2I, traffic lights can be made intelligent by tracing the number of cars passing by, then dynamically adjusting their timer clocks to change the traffic lights from green to red and vice versa, in response to traffic flow and density. V2I can also be used to summon for help during traffic accidents or for automatic vehicle toll collection. Most V2V communications are based on IEEE 802.11p standard while Cellular-V2X is a 3GPP LTE standard (approved in 2017) that supplements existing 5.9 GHz V2V communications (figure 10 and table 7).

Figure 10.

(a) V2X architecture and (b) vehicular platooning where vehicles in the rear follows the front vehicle.

Table 7.

Different types of vehicular communications and their features.

type	features and applications
V2I	examples are 802.11n, LTEtraffic flow regulationroad toll charging
V2V	ITS 5.9 GHz spectrum IEEE 802.11palert communications and short messageshazard warning propagationcollision avoidance
V2N	example is cellular LTE 4Gvehicle-to-network; requiring mobile network operator assistance to provide access to cloud-based data and exploit edge computing features
V2X	combinations of V2V, V2I, V2E (environment) [36–38], V2P (pedestrian), V2C (cloud), V2H (Home)refers to vehicle to everything (X)no cellular connection is used
Cellular-V2X [39]	3GPP Release 14 C-V2X standard, based on LTEfully compatible with 5G mobile technologiesprecise positioning, ranging and availability of traffic conditions for co-operative ITS and automated drivingsupports high density vehicle platooningsupports collision avoidanceprovides high bandwidth data communications

(h). Roads with smart intersections

As shown in figure 11, road intersections are prone to accidents due to road crossings and the obstruction of views of vehicles coming from different directions. Hence, work on smart intersection technologies and solutions [40] have evolved over the years. Cooperative intersection safety (INTERSAFE-2 2008–2011) [41] is a European project looking at these issues. The project uses sensors for vehicle and object detection, along with V2I, mapping and localization technologies to reduce fatal collisions and enhance safety at road intersections. The project provides a comprehensive accident analysis, identifying common accident scenarios and types at intersections that are most prevalent in several European countries. In that work, objects are classified into: (i) parked vehicles, (ii) moving vehicles, and (iii) pedestrians.

Figure 11.

(a) Accidents can happen at intersection in various ways and (b) blind spot in a four-way stop intersection.

In the VRUITS (ITS for Vulnerable Road Users) [42] intersection safety project, warnings are transmitted to road users (pedestrians, cyclists and drivers) with automated braking of car at intersections. In the smart intersection system [43] provided by Miovision Inc., intersections are able to sense and understand roadways situation and trigger real-time responses. Some of the responses include:

• (a)Extending green light periods to accommodate cyclists, so that they can make it through the intersection safely.
• (b)Sensing the presence of jaywalkers and warn drivers of connected cars approaching the traffic intersection.
• (c)Giving priority way of access to traffic lights for emergency vehicles.
• (d)Creating intelligence so that the intersection system can understand and analyse ‘near-misses’ from pedestrians, cyclists and drivers at intersections.
• (e)Reducing travel time by dynamic adjustment of traffic signal timing.

In the ‘smart intersection’ system proposed by Honda Inc., cameras are installed at each four corners of the traffic intersection, as shown in figure 12. Image processing software is then used to create a 360° image of the road intersection. Artificial intelligence and object recognition software are then used to classify objects into pedestrians, motorbikes, emergency vehicles, etc. Such information is then broadcast to vehicles at both roads forming the intersection. In this way, drivers' awareness of the intersection is greatly enhanced, overcoming blind spots and potential upcoming hazards.

Figure 12.

Example of intersection safety using cameras, object recognition, and V2X to warn drivers well in advance (source: from Honda Inc.—https://hondanews.com/) [44]. (Online version in colour.)

Table 8 shows the deployment of smart intersection technologies in USA and Japan. As commercial solutions become readily available, more countries are likely to implement smart intersections to reduce accidents and enhance safety of pedestrians.

Table 8.

Different types of vehicular communications and their features.

country	remarks
USA	Detroit, USA: In the City of Detroit, over 40% of intersections have been deployed with a smart traffic intersection system since June 2018.It has claimed to be the world's smartest intersection [43] with a system of sensors, video cameras, connected traffic signals and remote monitoring capabilities.The prime focus of the system is enhancing traffic intersection safetyThe system generates data pinpointing traffic-related fatalitiesArtificial intelligence is used to predict and avoid traffic hazards
Japan	In 2018, Honda announced a smart intersection technology for V2X designed to reduce accidents at traffic intersections.A demonstration of the system was successfully done at the City of Marysville, Ohio [44]The system allows drivers to virtually ‘see through’ and ‘see around’ buildings and wallsObject recognition software, intersection-mounted cameras and V2X communications are used.

(i). Emergency rescue architecture using V2X

Road accidents are still happening on a daily basis and 1.35 million people have died in road accidents each year [33]. Others are fatally injured as a result of the crash. In [35], the trend in traffic emergency services has evolved from using cellular calls to report an accident to using eCall, OnStar and the use of vehicular ad hoc networks (VANETs) over different wireless communication channels. After an accident has occurred, the critical hour is known as the ‘golden hour’ (figure 13a), where actions have to be taken quickly in order to save lives. More specifically, the ‘golden hour’ refers to the time elapsed between the accident and the arrival at the hospital. By reducing this time, injured persons will have a greater chance of saving their lives and reducing the severity of their injuries. Also, the use of vehicular communication capabilities can help rescue services and paramedics to speedily provide medical aid and rescue.

Figure 13.

(a) Golden hour in a car accident and (b) alert message sent by V2X system. (Online version in colour.)

In the current eCall method, an automatic crash notification alert is sent to the local emergency call centre whenever a vehicle has crashed. Call centres will then summon help and emergency services to the site of the accident. While effective, it is subject to delays and there is a lack of advance information about the severity of the accident and the conditions of those injured.

With VANETs [45], the alert ( figure 13b) sent out by the crashed cars can be used to alert nearby vehicles, alerting them to stop or slow down, and some people can render aid whenever possible (for example, one of the drivers or passengers could be a doctor, medic or fireman). The local wireless alert message propagation provides the fastest notification of the crash incident. Also, V2I will enable the crash alert to be propagated to the control centre, and to summon for emergency services. This can occur over 802.11p and via cellular 4G signals. Hence, the simultaneous alert and notification transmissions using V2V and V2I can improve the speed of getting help to the victims at the crash site.

Additionally, neighbouring vehicles can gather and transmit useful and accurate information about the accident, such as the crashed vehicle status and their occupants to the emergency centres. In this way, rescue centres will be able to correctly process and forward all the information to the respective agencies involved (such as the fire brigade, hospital emergency rooms, etc.). This information can be exploited by rescue team leaders for better planning of the rescue operation, before travelling to the accident site (for example: knowing in advance how to correctly extricate passengers from a specific EVs in order to avoid electric shock).

(j). Roads with smart street lights

(i). Technology and deployment

The first intelligent street light system was deployed in Oslo, Norway in 2006 (figure 14a). Its purpose was to control the on-off of street lights in order to save energy. Fast forward, some of the advances achieved today include:

• (a)connected street lights,
• (b)solar-powered street lights,
• (c)motion-activated street lights,
• (d)street lights as Wi-Fi access points,
• (e)data-analytics-enabled street lights, and
• (f)ambient-controlled street lights.

Figure 14.

(a) The first Oslo intelligent street light system in 2006 and (b) current smart street light. (Online version in colour.)

Most street lights today have now been replaced with LEDs instead of fluorescent or halogen bulbs for better energy efficiency, reduction in cost, ease of maintenance and improved operational control. Sensors and Wi-Fi are being added to street lights control unit to allow them to sense the presence of pedestrians and cars, hence turning on and off lights when needed (i.e. on-demand lighting). Wireless connectivity has allowed street lights to be connected, enabling them to form networks, and allowing them to be controlled remotely, operating in pairs or groups of lights [46]. As shown in figure 14b, sensors-added street lights can be used for a variety of purposes, such as

• (a)gunshots, terrorists and riots detection,
• (b)air quality monitoring,
• (c)EV charging points,
• (d)traffic congestion monitoring,
• (e)people crowd monitoring,
• (f)public safety monitoring,
• (g)roadside parking monitoring, and
• (h)trash and littering monitoring.

On another perspective, street lights can also serve as an advertising platform, through the use of attached digital signage on the street poles. However, too much advertisements on roads can be a distraction to drivers. As shown in table 9, city planners and leaders around the world have been embracing the use of smart street lights, as part of smart transport for the development of future smart cities (figure 15).

Table 9.

Smart street lights developments in UK, USA and China.

country	remarks
UK	Hampshire, UK—155 000 street lamps connected via Zigbee mesh networking. A wireless central management system controls on/off and dim/brightness of lights and monitors outages remotelyGlasgow, UK—Intelligent street lights connected via wireless mesh networks to save energy and enhance public safetyEdinburgh, UK—64 000 LED lights in the city are to be connected via Telensa's PLANet by 2020 to reduce energy and maintenance cost while allowing smart remote monitoring and control
USA	Cambridge, MA: City-wide LED street light retrofit, thereby achieving 80% reduction in energy usage. The ambient of street lights is automatically adjusted based on neighbourhood-specific profiles (activities, time-of-the-day, population, etc.)
China	Hongze County, Jiangsu Province: More than 3000 smart street lights were installed, along with attached equipment to offer Wi-Fi, environmental monitoring and digital signage servicesWeifang, Shandong: 40 000 smart street lights to be deployed using NB-IoT for connectivity

Figure 15.

Connectivity options, suitability and comparisons (source: Navigant Research). (Online version in colour.)

(ii). Smart street lights connectivity options and comparisons

As shown in figure 16, power line communications (PLC) and RF ad hoc mesh networks are possible connectivity options for connecting smart street lights. There is no mobility of street lights here and hence this is equivalent to a static ad hoc network. If one light is faulty, the neighbouring street light can be automatically brightened to compensate for the loss in ambience. This added ‘smartness’ into the system. Also, a maintenance alert signal can be sent to notify the respective city services department to have that faulty street light repaired or replaced.

Figure 16.

Smart street light solution from HUAWEI (source: https://www.huawei.com/minisite/iot/en/smart-lighting.html). (Online version in colour.)

Hybrid PLC and mesh ad hoc networks can be used to increase robustness and reliability. Cellular connectivity is viewed to be less attractive due to its high cost, the need for constant equipment upgrades and the risk of backward compatibility as new cellular communications standards evolve. However, NB-IoT solutions have been proposed [47] by Huawei Inc., for massive sensor connectivity and it can help to eliminate the difficult chores of street light network construction and maintenance by city planners since NB-IoT is an operator-controlled network. Essentially, telcos can aid the city government in these chores. Current companies that provide solutions and products for smart street lights includes Philips, Echelon, Telensa, GE lighting and Osram.

Of the 10 advances mentioned, musical roads, energy-harvesting roads and roads that weigh on-going cars have been deployed and in use. The wireless digital traffic sign is a totally new advance, along with the applications of V2X for wireless real-time traffic violation detection. All the remaining advances are in progress with some experimental deployments done in certain countries.

3. A peek into the future

Smart roads [48] will be an indispensable part of smart transport for future smart cities. In a period of over two decades, we have witnessed the progress and developments of various technologies that have helped the convergence and realization of smart roads. Figure 17 shows a snapshot of future smart roads that have included all the 10 technological advances mentioned in this paper. However, additional new advances may evolve from now that will further help reshape smart roads.

Figure 17.

Diagram showing future smart roads deploying the 10 advances mentioned in this paper. (Online version in colour.)

Smart roads will bring about greater automation, higher energy efficiency [49], lower cost, improved public safety, cleaner air, greener environment, less traffic congestion, fewer accidents and fatalities, and hence improving overall quality of lives for city residents [50].

Roads will no longer be viewed as static infrastructures but rather an ‘intelligent grid’, fully aware of the situation, context and the environment. Smart roads will also be ‘end-to-end’, meaning it will not only benefit the city residents but also the city leaders and operators. Residents will be able to interact with the smart transport entity through their mobile phones (as they are out on the streets, strolling or commuting), while city operators will be able to monitor traffic and environmental conditions remotely but attentively, responding in a timely manner on a demand basis.

In the future, there will be mergence of three major grids:

• (a)the information grid,
• (b)the electric grid, and
• (c)the transportation grid.

The three grids effectively blend and work together to yield a powerful construct of the future smart city, forming the city's ‘brain’, ‘nerves’ and ‘hands and legs’ that provide reach to places within and outside the city. The advances mentioned earlier support various converged and ultimate bigger goals. Figure 18 shows the ultimate goal being: (i) near-zero fatality, (ii) zero carbon emissions, and (iii) efficient logistics.

Figure 18.

Converged and ultimate goals of advances in smart transport. (Online version in colour.)

As smart roads progress towards realization, more research and development efforts will be needed in the area of personal, mobile and software applications (with the inclusion of artificial intelligence) to help citizens realize the full benefits of future smart transport. The future will have a greater inter-mix and embedment of ICT and transportation [51].

Furthermore, with many governments in different countries launching their smart cities and smart transport projects, it would not be long before one realizes the presence of smart roads. For bigger countries like USA, China and India, many big cities within these countries will have to be developed and transformed into smart cities [52]. Such mega efforts require considerable time, planning, organization and efforts. Many countries have established program offices to take charge of smart city developments. Cities with higher density of roads will have the highest penetration of smart roads. According to Wikipedia [53], USA tops the rank by having 6 733 024 km of road network size, followed by India (5 603 293 km) and China (4 859 500 km), respectively. Co-incidentally, China, India and USA are also the world top three most populated countries. Hence, these will be the countries leading the world in the transformation of smart roads.

4. Conclusion

In the paper, we have discussed the recent 10 technological advances and developments in the area of smart roads. They include: (i) energy-harvesting road, (ii) musical road, (iii) automatic-weighing road, (iv) electrified road, (v) roads with wireless digital traffic signs, (vi) roads with automatic traffic violation detection and notification, (vii) roads that talk (V2X), (viii) roads with smart intersections, (ix) roads with fast emergency rescue, and (x) roads with smart street lights. These advances will aid in the progress, development and realization of smart transport for future smart cities. Future roads will be made intelligent, sign free, safer and communicative. They will not only support the mobility of people and things but also generate electricity, feeding energy back to the energy grid. More advances will continue to evolve as new technologies and applications emerge, including the application of data analytics, deep learning and artificial intelligence technologies into smart transport. Ultimately, smart cities will progress towards the goal of attaining near-zero fatality and CO₂ emissions, transforming our lives for the better.

Data accessibility

This article has no additional data.

Authors' contributions

C.K.T is the first author and wrote 95% of this paper. J.A.S. drew some of the figures for this paper. J.C.C. reviewed and provided comments on this paper. F.J.M. reviewed and provided comments on this paper.

Competing interests

We declare we have no competing interests.

Funding

This work was partially supported by the Ministerio de Ciencia, Innovación y Universidades, Spain, under Grant RTI2018-096384-B-I00.