Fullscreen HTML5Med Res (??MB)HTML4

Computerised Traffic Management By Dr David Maddison, VK3DSM If you are frustrated sitting in traffic now, you may take solace in the fact that it could be far worse without computerised traffic management. Australia is a world leader in many of the traffic management systems Image source: https://unsplash.com/photos/aerial-photo-of-vehicles-on-highway-XICpU0Aulr0 described in this article. S treets and intersections with light traffic, such as suburban and country roads, generally do not require automated traffic flow control. They might instead use Give Way signs, Stop signs, or roundabouts to prevent traffic conflicts and keep traffic flowing. However, beyond a certain level of traffic flow, traffic lights are typically installed to control traffic better and prevent blockages. Contrary to popular belief, traffic lights are not always beneficial. Although traffic lights can reduce the likelihood of T-bone collisions, they can increase the likelihood of 14 Silicon Chip rear-end collisions. For example, the city of Philadelphia, USA, found that “… replacing (traffic) signals by multiway stop signs on one-way streets is associated with a reduction in crashes of approximately 24%”. Famous Dutch traffic engineer Hans Monderman said that stripping all traffic controls from a city resulted in safer roads (www.wired.com/2004/12/ traffic). I have also observed that traffic seems to flow more smoothly when a set of traffic lights is out of service. Regardless of the benefits or drawbacks of traffic lights, we are stuck with them. Given that, the best way to keep Australia's electronics magazine traffic moving is to coordinate them so drivers are not forced to stop at every intersection. There are levels of traffic management beyond that, intending to keep traffic flowing as fast and smoothly as possible across an entire road network. Examples of other traffic control strategies are variable speed limits, lane direction changes, ramp entry timings (metering), variable tolling and even changing the traffic direction of entire roads. As an example, some motorway onramps in Sydney’s North Shore change to offramps at certain times of day, depending on demand. siliconchip.com.au CNC - WATERJET CUTTING The world’s first desktop waterjet. standup. desktop. TILES Now cut anything with digital precision using high-pressure water COPPER GLASS STEEL ALUMINIUM WAZER is the first desktop water jet that cuts any hard or soft material with digital precision. The high velocity jet uses a combination of high pressure water and abrasive particles to cut through the work piece. With WAZER, we’re bringing this advanced technology to any size workshop. • Cuts Any Material: Metal, Stone, Glass, Ceramic, Composite, Plastic, Rubber & Foam • Compact size fits in any work space • Brings professional-grade fabrication into any workshop • Ideal for on demand cutting of custom parts • In-house capabilities reduces costly outsourcing • Great for prototyping, manufacturing, fine art, & instruction • Simple set up. Just connect to standard electricity, water & drain • Cold cutting. No heat. No fumes. No need for ventilation • Quickly go from design to cutting. Takes any DXF or SVG file • Assembled in the USA Wazer Specifications Cutting Table Capacity www.machineryhouse.com.au/register card 460mm Manual Rapid Traverse (X & Y Axis) Absolute Positioning Accuracy 1500m/mm 0.08mm Max Pump Pressure 4600psi Max Water Usage 1.9L/min CNC System WAZER Nesting Software WAZER WAM Drawing File .dxf, .svg Garnet Delivery 150g/min Garnet Hopper Capacity Material WAZER can cut through sheets of any material including metal, stone, glass, ceramic,composites, plastic, rubber, and foam. 13.5kg 2 x 240VAC 10amp plugs Power Order Code PRICE PRICE ex GST inc GST Desktop Standup W08720 W08720S $14,650 $16,250 $16,115 $17,875 View and purchase these items online: www.machineryhouse.com.au/SIC2404 SYDNEY BRISBANE MELBOURNE (03) 9212 4422 (08) 9373 9999 1/2 Windsor Rd, Northmead 625 Boundary Rd, Coopers Plains 4 Abbotts Rd, Dandenong 11 Valentine St, (02) 9890 9111 Specifications and prices are subject to change without notification. (07) 3715 2200 PERTH Kewdale ADELAIDEY OPENING JUL 03_SC_290424 Gift SIGN-UP 305mm Y Axis Travel Z Axis Travel Go from design to cutting, fast WAZER’s web-based software WAM is fast and easy. Load your design file into WAM and prepare your cut in minutes. There is no need for a dedicated PC, WAM is compatible with Chrome, Internet Explorer, Safari, Firefox web browsers. TO WIN PRIZES & RECEIVE DISCOUNTS 305 x 460mm X Axis Travel It is important to bear in mind that, no matter how advanced any traffic management system is, if there is insufficient physical road infrastructure to suit the demand, there will still be slowdowns or stoppages. However, the severity of those problems can sometimes be reduced with good management. This doesn’t just apply to cars, motorbikes, trucks and coaches. Bad traffic flow can also impede public transport vehicles such as buses and trams, as well as bicycle riders. Modern electronic traffic management comes within the purview of ‘Intelligent Transport Systems’ (ITS). ITS uses information and communications technologies, traffic sensors and software to manage a road transportation system. Cooperative ITS (C-ITS) involves road users communicating with each other, plus local and central traffic management systems, to improve safety and efficiency. Adaptive traffic control is a means by which traffic signal timing, variable speed limits, entry onto motorways (ramp metering), lane direction changes and other techniques can be varied to optimise traffic flows according to demand. Loosely speaking, a “platoon” of vehicles (or the French version of the word, peleton) is a group of vehicles travelling together. More strictly speaking, it refers to groups of vehicles travelling very closely together as part of an intelligent transportation system, with a controlled distance between them, much like the carriages of a train. If platooning is fully automated, it allows increased capacity on a given road, reduces air resistance (lowers fuel consumption) and reduces collisions, although it may result in lesser driver attention. Adaptive cruise control (where a radar or camera is used to keep a constant distance from the vehicle in front) enables a primitive form of platooning. A major objective of traffic signal coordination or “progression” is to allow a group or platoon of cars travelling along a particular route to get a highly desirable “green wave”, passing through many consecutive intersections without stopping. A motorist’s dream... Important parameters Before covering traffic management 16 Silicon Chip methods, we should explain what traffic flow parameters need monitoring and possibly adjustment. The basic variables to be dealt with and controlled as part of any traffic management strategy are as follows: • Volume or traffic flow – the number of vehicles passing a fixed point, measured in vehicles per hour. • Speed – vehicle speed, instantaneous or average, either individually or as a stream. It is typically measured in kilometres per hour (km/h). • Concentration or density – the number of vehicles occupying a given length of road at an instant in time, measured in vehicles per kilometre or metre. • Headway – the interval between vehicles passing a fixed point, measured in seconds per vehicle. • Spacing or distance headway – the distance between vehicles passing a certain fixed point, typically measured in metres per vehicle. • Occupancy – a measure of the proportion of time vehicles are stationary at a specific point in a lane, such as over a detector loop or other sensor. It is reported as a percentage. As part of any traffic management system, a wide variety of sensors must collect data like the above, to be analysed and acted upon in real time. They can use techniques such as: • A light beam that’s broken when a vehicle passes. • An inductive loop that detects the metal of a vehicle above. • Analysis of radar returns. • Image analysis from a camera. • Observations from within the traffic stream, such as via smartphones. • Active vehicle identification, such as at tolling points. • Other methods, which we will discuss later. Traffic signals (traffic lights) The modern traffic light or signal is the most fundamental form of electronic traffic management. The first traffic signals (for horse-drawn vehicles) were installed in London in 1868. They used semaphore-style indicators; the first electric traffic lights were installed in Cleveland, Ohio in 1912. The first four-way, three-colour traffic lights were installed in Detroit, Michigan in 1920. All of those were manually controlled, but by 1922, traffic lights were Australia's electronics magazine controlled electronically by automatic timers, saving vast amounts of labour. Australia’s first traffic lights were installed in Sydney in 1933 (see siliconchip.au/link/abu2). Additional traffic lights were not installed in Australia until 1937. Computers started to be used to control traffic lights in the 1950s. In 1963, Toronto, Canada implemented a computerised traffic light system that controlled intersections across the city with communication over leased phone lines, using 1000 vehicle detectors. That system was initially responsible for 500 intersections, with the control computer able to handle 1164 intersections. However, Toronto was rapidly expanding, and the computer was running out of capacity, so the system was upgraded in the early 1980s. Also in 1963, SCATS (Sydney Coordinated Adaptive Traffic System) started controlling eight intersections in Sydney’s CBD. We’ll come back to that system later. At the most basic level, traffic signals can be either fixed-time or actuated. Fixed-time or interval-based operation means the signals operate according to a fixed sequence and timing, repeating the same sequence regardless of the number of vehicles on each road. The timing of such signals may change according to a schedule. Such signals are now rarely used in Australia. Traffic-actuated or phase-based signals rely on the input from sensors, such as an inductive loop in the road, to determine timing and sequencing and adjust their operation according to traffic demand. That is for just one set of signals for one isolated, non-­ coordinated intersection. Beyond that, sets of traffic signals at multiple locations can be coordinated to facilitate the green wave. Isolated traffic-actuated and fixedtime controls are now not generally used in Australia since the increased adoption of SCATS Master Isolated (SMI) control. A SCATS controller may be run in its regular mode, linked to a regional computer as part of a wide-area traffic control system, or in an isolated mode for single non-coordinated intersections. When running a non-­ coordinated intersection, SMI still uses SCATSstyle adaptive algorithms. siliconchip.com.au Types of traffic light sequences include: • Fixed sequence that never varies. • Tr a f f i c - a c t u a t e d s k i p p e d sequence, where some sequences are skipped if there is no traffic needing a certain sequence. • Variable sequence that uses near real-time measurement of traffic via detectors to constantly adjust timing and sequences according to demand. • Priority sequence, inserting a sequence to accommodate a train, bus or tram. • Forced sequence, determined by a master controller in a centrally-­ coordinated system. There is also emergency vehicle priority or ‘preemption’ where traffic signal phasing can be adjusted to facilitate the passage of emergency vehicles, using techniques such as: • In-vehicle transponder. • Emergency sequences activated from stations or facilities near traffic signals to clear traffic. • GPS tracking of an emergency vehicle, communicated to a central controller to implement appropriate sequences as the vehicle arrives at each intersection. • A phone call to a central traffic control office to implement appropriate sequencing along the emergency vehicle route. Traffic signal coordination If you ever get a run of green lights (green wave), it is likely the result of traffic signal coordination to time the length & duration of the green phases. This concept is called “traffic signal progression”; it is “the practice of coordinating the operations of two or more signalised intersections” – see Fig.1. The problem gets more complicated the more sets of traffic lights are to be coordinated, and even more complex when considering crosswise traffic flows. Crosswise traffic might experience increased delays in such a system. Pedestrian movements and other road users also have to be considered. Traffic signal terminology Each possible legal trajectory of traffic at an intersection is called a movement. At a basic two-way intersection, there can be 12 movements, with traffic in each approach being able to go left, right or straight ahead. A traffic signal phase is a set of traffic signal indications applying to vehicles or pedestrians, allowing simultaneous non-conflicting legal movements. For example, a phase might Offset reference point is beginning of first co-ordinated phase yellow Fig.1: traffic lights can be synchronised to avoid vehicles having to stop and go constantly, but there are limits to such synchronisation. The purple and blue lines represent the outer limits of a green wave. Original source: www.kittelson. com/ideas/pros-and-cons-of-signal-coordination/ siliconchip.com.au Australia's electronics magazine May 2024  17 Fig.2: an example of an intersection with three phases and parallel pedestrian movement. Original source: https://austroads.com.au/ (Guide to Traffic Management Part 9, page 81). Fig.3: traffic light phases (intervals). Original source: https://austroads.com.au/ (Guide to Traffic Management Part 9, page 222). Fig.4: the desired sequence of operations for a four-way intersection, which can be implemented in a Programmable Logic Controller (PLC), microcontroller or other means, or in earlier times, electromechanically with relays and timers. have north-south traffic seeing green in both directions while also allowing left turns for both sets of traffic. A particular phase in the sequence can be skipped if there is no demand for it; eg, a right turn phase could be skipped if no cars are waiting to turn. Fig.2 illustrates an intersection with three phases. Phase sequence is the order of phases in a signal cycle. These may be fixed or altered according to demand. Signal groups are sets of individual lights that share the same colour and are all activated for a particular phase. They are identified by which phase they belong to, such as the green lights associated with phases A, B or C in Fig.2. A cycle is a complete rotation through all possible phases. The cycle time is the time taken to move through all possible phases (sequences) at an intersection. An interval refers to the change from one phase to another, either the running phase interval (green) or the clearance phase interval (yellow and some red) – see Fig.3. Phase split is the proportional of cycle time a given phase is displayed. Offsets are the time relationships between green phases of successive sets of signals when the system is coordinated. Vehicle detectors may obtain information for either ‘strategic’ or ‘tactical’ purposes. Strategic information is used to compute cycle length, phase splits and signal offsets. Tactical information is used to determine the demand or duration of phases. The traffic controller Original source: https:// instrumentationtools.com/plc-based4-way-traffic-light-control-system/ 18 Silicon Chip Australia's electronics magazine The traffic signal controller is the heart of a set of traffic signals. Older ones contained relays and mechanical timers, while modern types are microprocessor-controlled and receive inputs from various sources. They generate various outputs and communicate with other controllers and central management systems. Typical traffic signal sequencing is shown in Fig.4, while the inputs, outputs and communications for a typical controller are shown in Fig.5. One Australian company making traffic controllers is Aldridge Traffic Controllers, now owned by Siemens Mobility (siliconchip.au/link/abu1). They designed and manufactured the ATSC4 Adaptive Traffic Signal siliconchip.com.au Controller (Fig.6). It can manage up to 32 signal (phase) group displays with up to 64 inputs from vehicles, pedestrians or emergency services. It can operate in standalone mode or as part of an Intelligent Transportation System such as SCATS. It can communicate via Ethernet with a local network or 4G modem, ADSL or PSTN networks, supports VC6 SCATS protocol and DSRC (Dedicated Short Range Communications, see later) and comes with advanced software. Preventing hazardous signal combinations It would be disastrous if all signals at an intersection showed green simultaneously. This can be prevented by interlocked switching and/or conflict monitoring. For example, in relay-­ Fig.5: operation of a modern traffic signal controller. Original source: https:// controlled circuits, if one signal group austroads.com.au/ (Guide to Traffic Management Part 9, page 85). shows green, the conflicting signal ◀ Fig.6: the ATSC4 Adaptive Traffic groups are forced to red. Signal Controller, which is This can be done by methods like SCATS compatible. Source: www. cutting the power to conflicting green aldridgetrafficcontrollers.com.au/ signals when one is activated. In solid-­ products/traffic-signal-controllers/ state relay controlled systems, the outatsc4 puts must be monitored to ensure safe signal groups and avoid unsafe groups, as per Fig.7. Traffic sensors It is necessary to measure the traffic flow to control traffic. There are various ways of doing that. Stationary sensors can measure traffic flow, but in other cases, the data comes from vehicles. The latter example is known as Floating Car Data (FCD). FCD can also be used by Apps like Google Maps and Waze to provide information about road hazards such as accidents, construction works, potholes etc. Automatic Number Plate Recognition (ANPR) ANPR is used for tolling and legal compliance but can also provide traffic flow data. Optical character recognition is used along with algorithms to locate the position of the number plate in an image. Bicycle and pedestrian counters Some traffic management systems include bicycle counters. Bicycles are counted using much the same technology as cars. One example at Veloway 1 in Woolloongabba, Queensland is shown on the Department of Transport’s website (siliconchip.au/link/ abu3). It uses a camera and artificial siliconchip.com.au Fig.7: safe and unsafe combinations of signals. Original source: https:// austroads.com.au/ (Guide to Traffic Management Part 9, page 88). Australia's electronics magazine May 2024  19 Fig.8: a pedestrian counting system in the City of Melbourne. Source: www.pedestrian.melbourne.vic.gov.au/#date=26-022024&sensor=RMIT14_T&time=15 intelligence (AI) to classify traffic as either pedestrians, cyclists or riders of some other device. There is a Pedestrian Counting System in the City of Melbourne (see siliconchip.au/link/abtq). It uses laser or thermal sensors to record pedestrian movements. The sensors are connected to a 4G wireless data transmission system, a central server and a visualisation system. The data can be seen with an online visualisation tool at www.pedestrian.melbourne. vic.gov.au (see Fig.8). People-counting systems are also used for measuring occupancy in places such as shopping centres, entertainment venues, libraries, government buildings and retail stores. Cameras Software can be used to analyse video streams from any source to count and classify vehicle traffic. Such cameras typically use AI and machine learning (ML). An example is shown in Fig.9. Fig.9: analysing a video stream using the Camlytics software (https://camlytics. com/). Some software does offline analysis, like this one, while others do it in real-time. Source: https://camlytics.com/solutions/car-counting Fig.10: how an inductive loop traffic sensor works. Original source: www. researchgate.net/publication/287003681 20 Silicon Chip Australia's electronics magazine Mobile phone data (FCD) Tracking mobile phone signals from car users requires no roadside or other infrastructure, and nearly all cars have at least one mobile phone on board. Privacy concerns aside, no specific permission is required to do this, as mobile phone towers already obtain such data as part of their function. Location and speed data is obtained via triangulation of the phone signal and hand-over data from tower to tower. Collecting such data from large numbers of phones enables traffic flow to be monitored (and, incidentally, pedestrian traffic). Inductive loop sensors Inductive loop traffic sensors have an insulated wire loop or loops embedded in the roadway to detect traffic – see Fig.10. You can often see where they are because the road has been cut and resealed to embed the wire. The loop is energised at 10-200kHz. It acts as a tuned circuit that changes in frequency when a mass of metal, such as a car, is nearby. This change in frequency is detected by the associated electronics and interpreted as the presence of a vehicle. One problem with such loops is that they may not register the presence of a small vehicle such as a motorcycle, scooter or bicycle. We have also seen siliconchip.com.au cases where people stop short of the sensor and never get a green light! Figs.11 & 12: the TIRTL processor (left) and transmitter (below). Source: CEOS Pty Ltd. GPS data (FCD) Some phone apps like Google Maps and Waze (now owned by Google) upload GPS data, which is used for various purposes, such as choosing optimal routes to avoid traffic. In a sense, it is ‘crowdsourced’ traffic data. Infrared sensors An example of an infrared traffic sensor is the Australian-developed TIRTL (The Infra-Red Traffic Logger) – see Figs.11 & 12. It consists of an infrared transmitter and receiver on opposite sides of the road. As vehicles interrupt the beams, it can record the number and type of vehicles, their speed and which lane they are in. The information can be logged for statistical purposes or traffic control. It can also be connected to a red light and/or speed camera to record violations, detect over-height and overlength vehicles, be used for bus lane enforcement and various other applications. Some of its operating modes are shown in Fig.13. It is a product of CEOS (www.ceos. com.au/products/tirtl/) and is used in twenty countries. Commercial sales started in 2002. Fig.13: some of the operating modes of the TIRTL. Original source: www.ceos. com.au/products/tirtl/ Piezoelectric sensors Piezoelectric material converts stress into an electric charge, which can be measured to detect a load such as a vehicle. They can detect the number of vehicles, number of axles, vehicle speed and weight. Pneumatic road tubes These familiar devices, used on a temporary basis for traffic surveys, consist of one or more rubber tubes across a road. They sense vehicles as they drive over and compress air in the tube, activating a switch in the electronics box at the side of the road. Software can determine the number of axles and speed of the vehicle, plus the number of vehicles that pass. With two tubes, the travel direction can be sensed. Radar sensors Radar sensors measure road traffic and perform tasks such as counting and classification, incident detection, wrong-way detection, ramp metering, lane blockage detection and queue siliconchip.com.au Fig.14: lane-specific forward-looking radar detection using a smartmicro device (right) compared to a side-mounted radar device (left). Australia's electronics magazine May 2024  21 Fig.15: the smartmicro-MLR MultiLane Radar detector mounted on a pole. Source: www.yunextraffic.com/ wp-content/uploads/2023/06/YunexTraffic_Smartmicro-MLR_EN.pdf length measurement, among others. Objects such as pedestrians, bicycles, motorbikes, passenger cars, transporters, short trucks and long trucks can be sensed and classified. One such device from smartmicro (www.smartmicro.com) has multiple forward-firing beams, can simultaneously detect 256 vehicles and provide lane-specific detection for up to 12 lanes with a 500m range (see Fig.14). The device (Fig.15) can also be used to trigger speed and/or red light cameras. Detecting emergency vehicles A typical “emergency vehicle preemption” system involves an emitter attached to an emergency vehicle, a detector at a traffic signal and an optical signal processor. As an emergency or other priority vehicle approaches a signal, optical emissions are detected, and the signals switch to green for the emergency vehicle. A typical installation is shown in Fig.16, as per VicRoads specification TCS 055-1-2005. VicRoads uses the Tomar STROBECOM II emitter, detector and optical signal processor (see siliconchip.au/link/abtr). Fig.16: a typical emergency vehicle preemption system. Original source: www. vicroads.vic.gov.au/-/media/files/technical-documents-new/its-specificationstcs/specification-tcs-055--emergency-vehicle-preemption.ashx adjusting speed limits, changing freeway ramp entry timing etc. Products that do this include: Australian Integrated Multimodal EcoSystem (AIMES) AIMES is described as a “worldfirst living laboratory based on the streets of Melbourne, established to test highly integrated transport technology with a goal to deliver safer, cleaner and more sustainable urban transport outcomes”. It is an experimental system by the University of Melbourne, the Victorian Department of Transport and Planning, and industry partners. It uses a mesh of individual smart sensors to track pedestrians, cyclists and traffic within a city’s transport system of intersections, tramways, bus routes and traffic signals. The goal is to achieve more efficient and productive use of transport infrastructure. It is said to be the world’s first and largest ecosystem for testing new transport management technologies, incorporating 100km of roads bounded by Lygon & Hoddle Streets and Victoria & Alexandra Parades in Melbourne. Information from such a system could be used to operate a driverless car or improve pedestrian or cyclist safety. A 2.5km test corridor along Nicholson Street in inner Melbourne with comprehensive monitoring and sensors at every intersection provides improved traffic flow and safety for all types of vehicles and traffic. The Nicholson Street intelligent corridor integrates data from existing sources such as CCTV footage, Bluetooth signals from personal devices, the Sydney Coordinated Adaptive Traffic System (SCATS), General Transit Feed Specification (GTFS) and sensors specifically installed for AIMES – see Fig.17. Vehicle re-identification (FCD) Vehicles can be detected at one location and then at another location. This enables travel time and speed to be calculated between pairs or groups of sensors. A vehicle can be sensed by the MAC address of any Bluetooth device in the car, by reading RFID serial numbers from devices such as toll tags or using number plate recognition. Traffic management systems Once traffic data is collected, it needs to be analysed and appropriate actions taken. Possible actions include altering traffic signal timings, 22 Silicon Chip Fig.17: the “Kapsch Intelligence Corridor”, featuring part of the AIMES Nicholson Street “intelligent corridor” in Melbourne. Australia's electronics magazine siliconchip.com.au Machine learning and analysis are used to process CCTV images, then the EcoTrafiX platform is used to visualise and manage sensor data. Cloudbased AI and predictive models are also used. According to Dr Neema Nassir, the system uses “machine learning models that can optimise – through millions of simulation executions – the best right-of-way allocation, or the best green traffic light time allocation for competing modes and competing volumes”. ARCADY Assessment of Roundabout Capacity and Delay from the Transport Research Laboratory, UK (https://trl.co.uk/) is used to model roundabouts and “... predict capacities, queues, delays and accident risk at roundabouts”. COMPASS This traffic management system in Ontario, Canada, uses in-road traffic sensors to measure the speed and traffic flow on freeways. The data goes to a central computer so operators can view the data and cameras. They use the McMaster algorithm to change message signs and speed limits. Kapsch EcoTrafiX This traffic visualisation and management platform (www.kapsch.net/ en) is from Austria; see Fig.18. It includes traffic signal control, adaptive traffic control, event management, traffic prediction, travel information, data fusion and more. Other Kapsch products used in Australia and NZ are toll collection systems in Melbourne, Sydney, Brisbane and Auckland using Dedicated Short-Range Communications (DSRC), video-­based detection and classification and Automatic Number Plate Recognition (ANPR); the Nicholson Street intelligent corridor and AIMES and the Eastlink tolling system in Melbourne. In Queensland, they demonstrated a Cooperative Intelligent Transport System (C-ITS) to send warning messages about road works to appropriately equipped vehicles. MASSTR Meadowlands Adaptive Signal System for Traffic Reduction (www.njsea. com/transportation/masstr/) is an adaptive traffic control system used in the New Jersey (US) Meadowlands area, coordinating 125 traffic signals. It uses the Australian-designed SCATS software and is its fourth-­ l argest deployment worldwide. McMaster Algorithm This is a widely used traffic congestion detection algorithm based on the mathematical branch known as catastrophe theory. Speed, flow and lane occupancy (density) are analysed. If there is a dramatic loss in speed without a corresponding drop in flow and density, that suggests an incident has occurred. MOVA Microprocessor Optimised Vehicle Actuation from the Transport Research Laboratory (UK) was introduced in the 1980s for controlling isolated sets of traffic signals. NoTraffic NoTraffic (https://notraffic.tech/), founded in 2017 in Israel, is the world’s first AI-powered traffic management platform that fuses data from traffic sensors such as cameras, radar and information from vehicles via V2X (see later) and IoT technology with artificial intelligence. AI is used for NoTraffic’s computer vision neural networks and traffic optimisation algorithms. NoTraffic can be retrofitted at any intersection to connect it ‘to the cloud’. It can run in a fully autonomous mode, communicating with other intersections, road users and managers. Managers establish intersection and corridor policies with NoTraffic (see Fig.19). AI is used to classify and manage traffic according to those policies to maximise road capacity (see Fig.20). NoTraffic also provides information so traffic managers can better understand road networks by “understanding the root cause of traffic issues and applying the most relevant and effective solutions on a case-by-case basis”. NoTraffic can communicate with connected vehicles via V2X to provide alerts and rerouting information for accidents and hazards (see Fig.21). Fig.18: the Kapsch EcoTrafiX software. Source: NYSERDA Department of Transportation siliconchip.au/ link/abu4 siliconchip.com.au Australia's electronics magazine May 2024  23 It operates in Arizona and California, USA, among other places. One recently demonstrated capability of the system is the ability to detect a “red light runner” approaching an intersection and warn drivers with a green light going in other directions to stop to avoid a collision (see https:// youtu.be/aEuyUY28qzc). The NoTraffic video channel can be found at www.youtube.com/<at> NoTraffic Fig.19: NoTraffic allows intersection policies to be set up on its dashboard. OSCADY Optimised Signal Capacity and Delay is modelling software from the Traffic Research Laboratory (UK) that “calculates capacities, queues and delays for isolated (uncoordinated), traffic signal-controlled junctions. It can evaluate a set of known signal timings, and optionally, it can optimise stage (phase) lengths and/or cycle time to minimise delay”. PICADY Priority Intersection Capacity and Delay is modelling software from the Traffic Research Laboratory (UK) for the “prediction of capacities, queues, delays and accidents at isolated priority junctions”. Fig.20: NoTraffic uses AI to classify traffic types. Source: https://youtu.be/O_ Bpyuu_URI Rayven This Australian company offers an IoT platform for a “traffic monitoring and intelligent highway solution” to integrate “infield devices, sensors, third-party systems, and machinery to deliver real-time and predictive insights, as well as all-new capabilities to improve safety, maintenance, and use.” It is primarily for monitoring rather than traffic management – see siliconchip.au/link/abts Fig.21: information that NoTraffic might display in a V2X-connected vehicle. Source: https://youtu.be/O_Bpyuu_URI SCATS Sydney Coordinated Adaptive Traffic System was introduced in 1963 as a pilot controlling eight intersections in the Sydney CBD using valve-based IBM equipment. By 1970, DEC PDP11 computers controlled intersections, followed by microprocessor-based traffic signal controllers in 1974. SCATS is owned and developed by Roads & Maritime Services (RMS) in NSW. It is now used in many countries, controlling 37,000 intersections, and is considered one of the world’s leading adaptive traffic control systems. 4200 intersections in Sydney are controlled by one SCATS system. In Australia's electronics magazine siliconchip.com.au 24 Silicon Chip Victoria, SCATS controls over 4000 intersections in Melbourne, Ballarat, Bendigo, Traralgon, Geelong and Mildura. SCATS is used in another 150 cities in 27 countries, including the USA, Brazil, Singapore, India, Malaysia, Ireland, South Africa, Fiji and China. SCATS runs on Microsoft Windows via one or more regional controllers and a central manager computer. A central manager can control 64 regional controllers (regions). Each regional controller can manage 250 traffic signal controllers (intersections) for a total of 16,000 intersections. There is plenty of redundancy, as each regional controller can continue to operate even if communication with the central manager is lost. If regional controllers fail, there is a fall-back mode to local individual intersection control by the local traffic signal controller. The ATSC4 traffic signal controller is specifically designed to work with SCATS. SCATS controls three principal signal parameters: cycle time, phase split and offset. SCATS works at two levels: strategic and tactical. At the strategic level, regional controllers receive data from vehicle detectors to assess flow and occupancy data and optimise cycle length, phase splits and offsets for an area (groups of intersections). At the tactical level, individual traffic signal controllers use data from local vehicle sensors to omit signal phases if no vehicles are waiting. Even though there is tactical local control, ultimately, the system is coordinated by the regional controllers. SCATS uses a measurement known as the degree of saturation (DS), a measure of road capacity utilisation determined by traffic sensors during green phases. A figure over one means there is insufficient green time to satisfy demand and the road is congested. Cycle length is adjusted to keep DS around 0.9. Phase splits are also adjusted to keep the DS about equal for different approaches to the intersection. When using a SCATS traffic signal controller (eg, the ATSC4) for the first time, the signal controller will provide the initial default timings. Then the SCATS regional controller will start to adjust the timings (self-calibrate) according to the traffic flow at that junction. It will attempt to balance siliconchip.com.au and coordinate flows between neighbouring junctions as demand requires. SCATS can turn off coordination between intersections if necessary, such as during periods of light traffic when traffc at some intersections might flow better without coordination. SCATS can also prioritise the passage of certain vehicles, such as public transport buses and trams. If a pedestrian presses a button to cross the road, the signal phasing will be altered to run the pedestrian phase. SCATS can also be set for special events or other special purposes. VicRoads claims the following benefits from SCATS in Victoria: travel times down by 21%, stops down by 40%, fuel consumption down by 12% and fewer crashes due to smoother traffic flow. Fig.22 shows a SCATS interface window. The pie chart (on the left) shows the length of time for each phase, while a map of the junction is on the right, with the different signal phases shown to its left. SCOOT The Split Cycle Offset Optimisation Technique is an adaptive traffic control system for groups of traffic signals that are close together. It was first introduced by the Traffic Research Laboratory (UK) in 1979. Its purpose is to adjust signal timings based on input from sensors to minimise delays. It is used in 350 towns worldwide. SURTRAC Scalable Urban Traffic Control (https://miovision.com/surtrac/) was developed at Carnegie Mellon University in Pittsburgh, USA. It is an adaptive traffic control system that optimises traffic flows along corridors and complex urban grid networks. It uses artificial intelligence that treats the “intersection control challenge” as a “single machine scheduling problem” to optimise each intersection and share information with neighbouring intersections, to enable coordination and control across the whole network. The operational concept is: 1. Traffic conditions are established from sensor data. 2. The appropriate traffic signal phase schedule is computed for flow optimisation at intersections. 3. The schedule is transmitted to downstream intersections. 4. Rescheduling occurs every few seconds. It is used in Pittsburgh, USA and Peterborough, Canada. TRANSYT (TRAffic Network StudY Tool) This traffic modelling software was introduced in 1967 by the Transport Fig.22: a SCATS interface window. Source: www.aldridgetrafficcontrollers. com.au/ArticleDocuments/230/Introduction_To_New_Generation_Scats_6_5. pdf.aspx Australia's electronics magazine May 2024  25 Research Laboratory (UK) to optimise signal timing and perform simulations for “designing, evaluating and modelling everything from single isolated road junctions to large mixed signal-controlled and priority control traffic networks”. The results from the modelling can be used to optimise signal timing, SCOOT timings, for performance prediction and platoon modelling. UTC Urban Traffic Control, from the Traffic Research Laboratory (UK), takes data from SCOOT to coordinate traffic signal controls over a wide area, such as an urban road network. Veronet This traffic management system uses artificial intelligence to manage traffic and traffic signals, supporting inflows and outflows for a city, and optimising certain directions. It also supports autonomous driving modes for cars that support that mode. See www.veronet.eu/home.html Waze This navigation software (Fig.23) is “free” (because your data is the product). It is now owned by Google. It collects vast amounts of user data for driver navigation and other purposes. However, traffic managers can also feed that data to traffic management software or use it to visualise traffic flows, monitor conditions on key routes and observe changes over time. Heavy vehicle monitoring in Australia The National Heavy Vehicle Regulator operates a network of fixed digital cameras called the National Safety Camera Network to monitor the movement of heavy vehicles by recording number plate data. The network has over 120 cameras covering more than 5800km of road across five jurisdictions with an average of 4.2 million “sightings” per month. According to their website, they “use safety camera, registration, crash, defect, intercept and infringement data to generate profiling reports to identify operators, vehicles, drivers and infrastructure of interest”. There are also five mobile Automatic Number Plate Recognition (ANPR) cameras to detect the number plates of passing heavy vehicles. Their website shows trailer-, vehicle- and dronemounted cameras. According to the website, the mobile cameras are “used to develop policies and programs to increase road safety”. Traffic Management Channel The Traffic Management Channel (TMC) is a worldwide system delivering digital traffic data via commercial FM broadcast stations that can be displayed on a car’s built-in GPS map system (or, in some cases, add-on systems). It is incorporated into the existing Radio Data System (RDS), typically used to transmit station identification and program information. The protocols used for RDS-TMC data are ALERT C or TPEG. Such data can also be delivered via Digital Audio Broadcasting (DAB) or satellite radio. Information that can be delivered relates to traffic events, containing an event code, location code, expected incident duration, affected extent and any other relevant details. Vehicle navigation systems can use this data to generate an optimum route. This system only requires the reception of an FM, DAB or satellite radio signal from a cooperating broadcaster. Intelematics Australia (www. intelematics.com, owned by the RACV) broadcasts encrypted RDSTMC data under the brand SUNA Live Traffic to provide live traffic updates to participating in-car navigation systems and compatible add-on GPS devices. Originally, SUNA was only transmitted via FM radio, but today, it is also delivered over the mobile data network. According to the peak body for advanced transport technology, ITS Australia (https://its-australia.com. au/), SUNA is used by 90% of vehicles in Australia and NZ. According to ITS, their “road traffic data is collected through thousands of probes and sensors located on roads, in vehicles and infrastructure” and “We enrich our data using multiple proprietary sources and machine taught algorithms”. Intelematics has advised us that SUNA will be discontinued. In the past, Intelematics also maintained historical traffic databases that could be used for future road and traffic planning via the discontinued software tool INSIGHT (siliconchip. au/link/abtt). The INSIGHT software tool allowed visualisation of historic and present real-time data of such parameters as traffic volume or turning volume at intersections over periods of 15 minutes, days, months or years. It allowed the impact of various events or infrastructure changes to be determined. VicRoads “Smarter Roads” Fig.23: Waze data being used to manage traffic by the Port Authority of New York and New Jersey. Source: https://support.google.com/waze/partners/ answer/10715145?hl=en The VicRoads Smarter Roads program (see siliconchip.au/link/abu5) includes CCTV, travel time sensors, live travel information signs and pedestrian Australia's electronics magazine siliconchip.com.au 26 Silicon Chip crossing sensors. There are 2500 CCTV cameras covering most suburban traffic signals in Victoria, used by the Traffics Operations Centre to monitor traffic incidents and traffic flows. There are also 400 wireless travel time sensors and 43 live travel time signs. According to VicRoads, these cameras are not used for law enforcement purposes, and the video is not recorded, so it is not available for evidentiary purposes, such as for accident liability. However, that could change. Pedestrian detectors determine the number of pedestrians waiting to cross the road & prevent unnecessary waiting (https://youtu.be/vyyN92qT6OY). They also monitor roadside air quality. Fig.24: Sydney’s WestConnex road and tunnel network use Smart Motorways technologies. Source: www.westconnex.com.au/explore-westconnex/ WestConnex Smart Motorways WestConnex private motorways around Sydney (Fig.24) use “Smart Motorways”, their proprietary name, for technologies such as vehicle detection, CCTV cameras, ramp signalling, lane use management and variable speed limits. Smart Motorways are designed to operate and integrate with the rest of the Sydney (non-WestConnex) road network and the existing SCATS system. WestConnex Motorway operations are controlled from the Motorway Control Centre (MCC) shown in Fig.25. Self-driving vehicles Australia’s laws do not currently support autonomous vehicles on public roads; the National Transport Commission released a policy paper on the subject in 2022. An Automated Vehicle Safety Law (AVSL) is proposed by 2026 (siliconchip.au/link/abtu). In the USA, California allowed driverless taxis in San Francisco, but permission was suspended after an accident with a pedestrian. Similar laws are under development in several countries. V2X V2X or “vehicle-to-everything” refers to communication to and from a vehicle for traffic management and other purposes. V2X incorporates concepts such as those listed below and shown in Fig.26: • V2D (vehicle-to-device): Apple CarPlay or Google Android Auto. • V2G (vehicle-to-grid): connecting an EV to a smart electrical grid. • V2I (vehicle-to-infrastructure): a siliconchip.com.au Fig.25: Australia’s largest Motorway Control Centre (MCC) at St Peters, Sydney, with 60 panels. It provides monitoring and incident response for the M4, M8 and M5 East motorways. Source: www.westconnex.com.au/media-releases/ australia-s-largest-motorway-control-centre-supporting-westconnex-motorists/ Braess’ Paradox Braess’ Paradox is the counter-intuitive idea that adding an extra road can increase the average travel time. Conversely, closing roads can sometimes decrease travel time (of course, that isn’t always true!). The idea is used in traffic planning and management. For example, a section of road could be opened or closed depending on traffic conditions. The basic problem is that drivers don’t know what other drivers are going to do. If a new, high-capacity road is opened, many drivers who would otherwise take different routes might decide to use that road, resulting in their paths intersecting and generating heavy traffic and delays. If a smaller number of the drivers took the new road while others remained on the smaller roads, the average travel time could decrease, but that would require either good luck or coordination. It is also applicable in electrical networks, biological networks and even sports; for example, the addition of a champion player might decrease the team’s overall efficiency if there is an over-reliance on that player. For more information, see the video on “The Spring Paradox” at https:// youtu.be/Cg73j3QYRJc Australia's electronics magazine May 2024  27 vehicle communicating with traffic lights, parking meters etc. • V2N (vehicle-to-network): comms via WiFi or the mobile network for remote diagnostics and monitoring. • V2P (vehicle-to-pedestrian): provide alerts from vehicles to pedestrians’ smartphones, coordination with pedestrian crossings, prediction of pedestrian behaviour, automatic sounding of vehicle horn. • V2V (vehicle-to-vehicle): exchanging data with neighbouring vehicles, such as warning of Fig.26: some examples of V2X communications in a country that drives on the right-hand side of the road. Original source: www.researchgate.net/ publication/279765559 vehicles or pedestrians that cannot be seen directly due to obstacles, or an approaching emergency vehicle. Information for V2X can be obtained from various sensors, as shown in Fig.27. Sensor data management and communication are performed by the V2X OBU (On-Board Unit). An example of a commercial OBU is shown in Fig.28. The original V2X technology was based on WLAN (Wireless LAN) IEEE 802.11p, which is now incorporated into IEEE 802.11. The term used by the SAE (Society of Automotive Engineers) for this technology is DSRC (Dedicated Short Range Communication). In Europe, it is known as ITS G5. DSRC has a range of up to about 1km, supporting V2I and V2X. Unfortunately, the DSRC systems used in Europe, Japan and the USA are incompatible. In Australia, DSRC uses the 5.9GHz band. Australian E-Toll tags use RFID transponders with a DSRC protocol. DSRC can also be used for cooperative cruise control, cooperative collision warning, warning of an approaching emergency vehicle and warning of a railway level crossing. 3GPP C-V2X uses mobile networks for V2X communications. C-V2X also uses the 5.9GHz band, like DSRC, for short-range communications and has about 25% better range than DSRC. There is no restriction on range as long as a mobile tower is nearby. It supports V2I, V2V and V2N. It was originally based on 3G but now uses 5G. DSRC and C-2VX are competing technologies. Variable tolling Fig.27: vehicle sensors that might be used for V2X communications and other purposes. OBU stands for On-Board Unit. Original source: www.researchgate. net/publication/279765559 Some authorities advocate variable tolling, supposedly to reduce congestion, as a form of traffic management. Such a scheme operates on the Sydney Harbour Bridge. According to Wikipedia, it has been minimally effective, only reducing traffic by 0.19%. Other management systems Fig.28: a Commsignia ITS-OB4 V2X on-board unit (OBU). An equivalent roadside unit can receive data from units like this for traffic management purposes. Source: www.itsinternational.com/its2/products/ commsignia-gets-green-light-c-v2x-units 28 Silicon Chip Australia's electronics magazine Traffic management systems aren’t just restricted to roads. Systems are needed for air traffic management, space traffic management (to ensure satellites do not collide), rail traffic management, sea and harbour traffic management and even underwater traffic management! Similar schemes and approaches apply, but generally with different sensors. SC siliconchip.com.au