Signal of Intent

There have been many developments in Traffic Signals over the last ten years. This period has arguably seen the greatest amount of developments in the history of the industry, with many new technologies and solutions being widely deployed and adopted. However, with the current pace of change being experienced both directly in traffic systems and in newly emerging fields, such as autonomous vehicles, the next ten years looks set to result in far more fundamental changes.

In recent years many technologies becoming far more energy efficient. An example of this can be seen in traffic signals, where the Halogen lamps which had been used for the previous thirty years, have largely been replaced by Light Emitting Diodes (LED’s). However, the process has not been straight forward. To achieve a bright display, early products tended to use a disc full of LED’s, yet even with the quantity of LED’s used, it was still common to resort to overdriving them in order to achieve the required illumination levels. This resulted in premature failures and the widespread industry impression that the technology was not rugged enough for traffic signal use. Nevertheless, advancements in LED technology have since provided capabilities that could only have been dreamed of fifteen or so years ago. Signals now often use a central light source which is akin to a traditional ‘bulb’, but made using the latest generation of LED technology. These have resulted in impressive energy savings and are now so reliable that the traditional practice of bulk-lamp changes (where all the ‘bulbs’ would be routinely changed once or twice a year), is no longer necessary. LED’s are likely to remain as the de-facto technology used with signals in the future.

Over the last decade, the use of Extra Low Voltage (ELV) equipment has become widespread. Traditional signals used mains electricity, leading to safety concerns, both in the event of an incident (such as vandalism or when an errant vehicle strikes a pole), or whilst maintenance procedures are carried out. ELV systems have overcome these fears by using equipment which operates at less than 50v, thereby minimising the severity of electric shocks if they do occur. The adoption of this technology required a solution to the issues thrown up by the lamp monitoring requirements (where signals are actively checked to determine if a ‘bulb’ has failed), before it could be used reliably. This was eventually solved by using LED signals.

In another safety related area, Passively Safe street furniture has become commonly used, especially on high-speed roads. The idea of Passive Safety is to provide a forgiving roadside, where in the event of driver error, the risk of serious injuries occurring is minimised. This is achieved by ensuring that if a vehicle strikes a piece of street furniture (such as a sign or traffic signal pole), that it will yield or breakaway in a prescribed manner which minimises the energy transfer from the vehicle to the roadside structure. The result is that regardless of why a vehicle may have ended up striking a roadside feature in the first place, the driver will still have the opportunity to use steering and brakes to bring the vehicle to a halt in a controlled manner, and the vehicle occupants should not have been exposed to excessive forces. Passive Safety has raised technical challenges when used with traffic signals, because of the requirements that street furniture with electrical supplies must include an automatic isolation facility to turn off the electricity in the event of a vehicle strike occurring. The best solution for achieving this is dependent on the failure mode of the roadside structure being used, although the package of requirements and solutions is now widely understood and has seen the gradual increase in the use of passive installations.

One area which has benefited from a range of technological developments is detection. Within traffic control, the need to be able to detect vehicles, cyclists and pedestrians reliably has seen many innovations over the years. Since the 1960’s and 70’s, Inductive Loops have been widely used to detect vehicles. These work in a similar fashion to a handheld metal detector, but use a coil of wire cut into the road surface, to detect vehicles moving over them. Inductive loops are still widely used, and are regarded as a benchmark against which the performance of other technologies are measured. However, loops are susceptible to damage, and are expensive to renew. Because of this, other solutions have been pursued, including above ground detection technologies. From the basic microwave and infra-red technologies which were used to detect moving or stationary vehicles respectively, the capabilities of units have progressed over recent years. Microwave units have been developed to dramatically improve their accuracy with some having capabilities to detect multiple or stationary targets. Units which incorporate ranging capabilities now offer a seamless detection methodology for approaching vehicles (instead of the series of loops on the approach to signals used with traditional vehicle activated modes) and has seen the demise of infra-red detectors used for presence detection. These have also been used for motorway incident and flow detection, although initial results have been compromised by obscuration issues, where tall vehicles in the near-side lanes block sight of the outer lanes. This does raise questions about the need for achieving 100% detection rates, particularly if this can be supplemented from other sources, such as floating car data which is commonly derived from mobile phones in vehicles.

In addition to microwave, the use of optically based detection methods, including thermal imaging techniques, has improved the reliability of pedestrian detection (as used for kerb-side detection at PUFFIN crossings) and can provide a reliable means of differentiating cyclists in mixed traffic conditions. Although not commonly used in the UK for traffic signals, video based detection has seen massive improvements in capabilities over the last few years. Examples include the ability to include multiple detection zones and ‘self-learning’ capabilities such as anomaly detection (where the system can flag up abnormal activities, such as vehicles driving in the wrong direction). A major area of development in the field of detection in the past ten years has been the introduction of magnetometers. These solid state devices detect anomalies in the earth’s magnetic field, which when used for vehicle detection are caused by vehicles passing over the units. The first units used a small ‘puck’ shaped unit with a magnetometer device, wireless communications facility and a battery pack in them. These are installed by being core drilled into the road surface and communicated with roadside beacons which were connected into the roadside cabinet. However, the life expectancy of these units is ultimately limited because they are battery powered. Despite early issues with achieving reliable operation, the requirements of these units are now widely understood, and the dependability of the communications and batteries has also improved. To overcome these issues though, a new sub-genre of magnetometers has emerged which use a cable connection for power and communication. Magnetometers are now being widely used, not only for traffic counting and signal control, but also for Smart City applications such as parking bay occupancy detection (where drivers using an app can find free parking spaces more easily). In coming years, the complexity of vehicle detection at signalised junctions may be dramatically reduced (to just using presence detection at each stop-line), if low-cost distributed detector nodes are used to replace or supplement SCOOT type vehicle detection networks.  

Although signalised pedestrian crossings would appear to be a straight forward, the issues caused by the plethora of different signalised crossing types are ongoing. For many years, there has been an aspiration to move towards using near-sided pedestrian displays at signalised crossings (instead of the far-side Red and Green figure displays which people are used to seeing at Pelican crossings), to improve the safety and efficiency of crossing facilities. However, although the 2016 update to The Traffic Signs Regulations and General Directions (TSRGD) has stopped new Pelican crossings from being installed (in an attempt to gradually phase out installations with the flashing amber display in the vehicle signal sequence), the continued use of far-sided pedestrian displays at junctions (where the pedestrian display sequence utilises a blackout period after the green figure has extinguished) will now also be permitted at stand-alone crossings. There are currently more than a dozen permitted signalised crossing variants in the UK which leads to confusion by both pedestrians and drivers about how these installations operate. A recent addition to crossing facilities is the use of pedestrian count-down timers. These operate after the green figure and display the remaining period before the conflicting traffic phases start to run, providing pedestrians with additional information. These are popular with users because of the reassurance they offer, but can only be used reliably where the signals are operating under a fixed time regime. It would be good if over the next few years’ progress is made to rationalise signalised crossing types, possibly to a single family using near-side pedestrian indicators supplemented with additional displays such as count down timers.

Another change in TSRGD 2016 saw the introduction of small cycle signals mounted on the primary signal pole at junctions. These provide cyclists with a pre-signal allowing them to establish themselves before the following mixed traffic is allowed to move forward. The signals themselves are similar to the secondary traffic signals used in countries like France, and have a ‘cycle’ mask on each aspect to clearly identify them. This does raise the question that at some point in the future, the use of this type of continental display might be permitted instead of the current vehicle secondary displays, which are normally located on the far-side of the junction but can cause confusion for foreign drivers. Using the continental displays could offer a cost saving for new installations as they would reduce the number of poles required. 

Communications is an area which has seen dramatic changes over the last few years. Most remote traffic infrastructure which had a communications facility tended to use a telephone line, often using a permanently connected leased line. This provided a capability to control items such as traffic signals in real time, by using a simple bit pattern interface to control and confirm the current status. However, in recent years, there has been a large movement over to Internet Protocol (IP) based communications. This has largely been led by the withdrawal of leased lines by BT in the UK, but also because of a desire to use more data intensive applications. A diverse range of technologies are now being used for the provision of communications to traffic systems, including fibre-optic, ADSL, wireless MESH and 4G mobile communications, depending on the required functionality, location and budget. Because of this, it is now possible to control equipment in real-time, while also using other facilities, such as interrogating sub-system settings and connecting to additional equipment, for example CCTV units. It is probable, that traffic installations will become data nodes for a variety of highway based technologies over the coming years, due to their connected status and distributed nature across the road network. In Chicago, the city has been installing what it has called the ‘Array of Things’, which consists of diverse sensors mounted in compact pole mounted housings, typically located at traffic signals. These will provide both the city and citizens with detailed information across a range of parameters, including traffic, weather, pollution, flooding, noise and lighting conditions.

Traffic signals operate by using a controller located within a roadside cabinet adjacent to the junction. This is normally connected to each traffic signal pole with thick multi-core cables, which individually switch each individual signal aspect from inside the cabinet. The need to terminate a large number of cables in the controller has always determined the footprint needed within the cabinet, resulting in them being quite large and difficult to accommodate in cramped urban settings. However, in recent years an alternative way in which to control the signals has been developed, which uses a common power and data ‘ring-main’ connected to all the signal poles on a site, which power and communicate with a series of control nodes on each pole. This type of methodology will dramatically reduce the extent of the underground ducting normally required to fit all the necessary cables, leading to significant savings in construction costs for new installations. This development may eventually lead to a distributed topography for each element of equipment, removing or at least minimising the requirement for the traffic signal controller cabinet. In addition, the move away from dedicated server based Urban Traffic Control (UTC) systems located in local highway authority control rooms, to using cloud based virtual servers will see increased opportunities to keep abreast of technological developments and to work cooperatively across jurisdictions, modes of transport and even with non-transport related stakeholders. The use of Urban Traffic Management & Control (UTMC) type systems allows for a vast range of disparate sub-systems (including legacy equipment) to be integrated into a homogenised environment, which far exceeds the operational capabilities of the separate sub-systems on their own. 

The emergence of autonomous vehicles is widely heralded to occur in the coming years. Many have predicted that this will lead to the demise of traffic control systems, while others have promoted the idea of removing traffic signals altogether. To achieve this would require an ambitious road building programme particularly in our urban centres (to segregate different modes of transport, such as pedestrians and cyclists), however it is unlikely that the public has the appetite to allow the largescale construction of flyovers and underpasses following the blight that was caused in many cities by similar schemes in the past. As well as the necessity to ensure pedestrian safety, the requirement to retain traffic signals for the foreseeable future will be caused by the existing vehicle fleet. The average age of vehicles in the UK has been more than seven years for some time now, so traditional non-autonomous vehicles are likely to remain a significant proportion of the vehicle fleet for many years to come. Although the form of ITS infrastructure will change, the requirements to retain much of the traditional ‘heavy’ infrastructure will therefore remain over the next decade. In all likelihood, the requirements for information about the road network will increase, to ensure the full capabilities of autonomy are achieved in the safest and most efficient manner possible. The ubiquity of mobile communications will undoubtedly revolutionise the industry. As well as providing the basis of Vehicle to Vehicle and Infrastructure (V2X) communications (which will be a major step in the autonomy pathway), it could allow new data sources, such as probe data, incident detection and meteorological information to be gathered and used to control and inform across different modes of transport. Furthermore, the use of ‘light’ infrastructure based on Internet of Things (IoT) principles will see the deployment of low cost sensors to augment and replace some elements of the current infrastructure and to infill in areas which are currently poorly served (such as rural roads). This will have the capability to deliver data for a range of parameters at the ‘hyper-local’ level, providing a granularity which is currently unachievable. The continued miniaturisation of electronics (with rapidly increasing processing capabilities, reduced power consumption and improved wireless communications), will allow far more affordable ‘off-grid’ solutions to be realised. The ‘richness’ of available data will therefore increase dramatically over the coming decade, opening the opportunity to readily provide currently unimagined facilities.  

The need for traffic control will therefore remain, even with the advent of autonomous vehicles. However, its appearance is likely to change and its features will undoubtedly become far more complex. The requirements for the provision of increasingly diverse and feature rich data sources will play an important part in the autonomy pathway. The goal should be to ensure that the general public’s daily lives are enriched by the capabilities achieved.

Brought to you by ITSnow.org | The Home of ITS Online.

This article was first published in the December 2016 edition of Thinking Highways Magazine.