Future Trends

The Vehicle-to-Infrastructure (V2I) initiatives that are active in the US, Europe and Japan provide the potential for all vehicles to perform probe functions for private sector service providers and public traffic management authorities. The challenge will be dealing with all of this information, captured over an ever larger area. Innovations in visualisation tools such as Artificial Intelligence (AI) and Knowledge-Based Expert Systems (KBEST) are already being commercialised and provide increasingly valuable assistance to TCC staff in monitoring the state of the network – whatever the source of the data.

The combination of Vehicle to Vehicle (V2V) and Vehicle to Infrastructure (V2I) means that road users will be able to share their own sensor data (such as speed, grip, surface temperature or queuing) with a centralised computer facility for interpretation or direct transmission to other vehicles. A vehicle without sensors may still be able to transmit road condition status gathered through in-vehicle monitoring – based on simple indicators such as use of fog lamps to indicate fog and extended use of windscreen wipers to indicate heavy rain. Data collated from enough vehicles (with or without sensors) could enable an accurate picture to be produced of flow rates, speed and driving conditions.

Stakeholders in traffic data collection include:

• the road authorities and private companies that operate roads
• police and enforcement agencies with CCTV and other detection equipment
• vehicle fleet operators such as freight haulage firms, taxi companies or a public bus operator’s probe vehicle data
• private sector organisations that own their own sensor networks (such as a University’s weather monitoring equipment)
• crowd-sourced information providers

New opportunities for road traffic data capture and consumption between road users emphasise the need for cooperation between organisations on data sharing. (See Other Sources) A structured model for traffic data collection and interpretation - where the interfaces and responsibilities are well-defined - is being supplemented by new sources of data and analysis exchanged between users and third parties, where the TCC has little or no influence on the quality of data or its use.

Social network messaging and navigation service providers already provide real-time traffic data to road users, specific to the user’s location, based on crowd-sourced data. These sources can improve the effectiveness of a traffic and incident management operation, at a lower cost than direct investment in measurement systems. This is particularly useful for a road operator:

• which does not already have an extensive network of static sensors
• or who wishes to improve the level of detail of an existing coverage area at a lower cost than direct investment in measurement systems

Whilst crowd-sourced data can provide real-time traffic information at low cost - the level of detail depends on the density of numbers of mobile sensors. Less information will be provided during off-peak periods and where the proportion of the crowd that provides useful data is low. Road operators also need to be aware that the algorithms developed by third parties to meet commercial objectives may not be consistent with the approaches required for traffic management or for incident detection.

The road operator needs to bear in mind that it has a responsibility to all road users - not only those that have access to advanced mobile phones or vehicles equipped with on-board sensors.

Static roadside sensors are becoming more accurate and are available at lower cost. Improved methods of planning are being developed to make best use of available budgets. They recognise that the purpose of any detection scheme – and the reliability of individual detectors – will influence decisions on where detectors are deployed. For example, the optimal planning strategy for journey time monitoring would require detectors to be distributed widely - whereas flow measurement would demand a greater concentration of detectors to ensure sufficient coverage with some redundancy (to provide resilience in the event of equipment failure).

The investment in deploying new sensor networks is driving the development of combined sensors. For instance:

• the measurement of harmful emissions with vehicle counts within compact roadside enclosures, as shown below
• programmable bidirectional road studs (‘cats eyes’), as illustrated below coupled with external sensors to enable hazards to be reported directly to drivers through coloured LEDs
• road studs that can be interfaced with local condition monitoring systems to alert the driver to hazards or simply provide illumination when triggered by vehicle presence detection (the figures below are examples of both)

Transferring information from these types of devices requires some form of low power communications and development of Mobile Ad-hoc Networks (MANETS). This will enable sensors to transfer data between themselves over short distances and to fixed roadside data receivers. In dense urban environments, a network of vehicle or environment sensors can be installed cost effectively.

Extending this further into the future, combining MANETS with miniature sensors and some means of ‘harvesting’ energy from the environment may result, one day, in ‘smart dust’. This is an emerging technology comprising a system of tiny Micro-Electromechanical Systems (MEMS) that can detect light, vibration, temperature, magnetism or chemicals. In the future they be embedded in materials such as the white paint used to indicate traffic lanes.

These types of pervasive sensing and communication (via fixed roadside wireless receivers) can further improve the visibility that the Traffic Control Centre (TCC) has of local traffic conditions.

The increasing variety and geographic coverage of static and mobile sensors presents TCCs with major challenges in interpreting data and making rapid decisions on the appropriate response. This is particularly the case when traffic conditions change quickly because of an incident. Artificial Intelligence (AI) is already being trialled to optimise traffic signal timings, to make better use of available road capacity and to reduce the environmental impact of traffic. (See Intelligence in Transport)

Knowledge-based Expert Systems are able to maintain reference models of normal traffic conditions anywhere on the road network and can provide immediate recommendations on alternative traffic plans in the event of an incident. These are being trialled in London and are planned to be used in Hong Kong as part of its Traffic & Incident Management (TIM) System (See PDF 3202 Case Study Traffic & Information Management (TIM) System, Transport Department (Hong Kong))

Other trends include:

• integration of Geographical Information Systems (GIS) that improve the TCC operator’s view of the road network in 3-dimensions. An operator could, for example, ‘fly’ along a road to check traffic signs and visualise the status of any roadside sensor with an overlay of real-time traffic conditions interpreted from the sensor data;
• the development of the Communications Access for Land Mobiles (CALM) - or similar standards that define many of the communication systems - could enable Vehicle to Vehicle (V2V) and Vehicle to Infrastructure (V2I) communications;
• smart phones and novel display devices provide a readily accessible platform for applications aimed at informing users of real-time road and environmental conditions (using data captured by public authorities or private sector service providers).