Traffic and Transportation Telematics

The background

Urban Traffic Control Systems in the UK have been introduced in late 1970s. The SCOOT system (Split-Cycle-Offset optimisation system), developed by the Transport Research Laboratory (TRL) underwent several upgrades and is now capable of monitoring and controlling traffic networks with hundreds of signals while its original objective of reducing delays, congestion and stops remained unchanged. However, with road networks reaching saturation and a recognition that commuters must be persuaded to abandon their cars in favour of public transport, one can conclude that the best days of centralised UTC are over.
 



Instead the power and versatility of basic traffic control systems is now being incorporated into an evolving comprehensive Demand Management Toolkit, balancing the need to provide bus priority and maintain air quality with the instinctive desire to maximise traffic flow. The research work in the ROMANSE project in Hampshire and the parallel development of the UTMC system in Nottingham are demonstrating the added value to be gained by linking UTC with other systems such as vehicle location and passenger information schemes.

The ability to share large amounts of data generated by ITS will provide the best opportunity to ensure that travellers have access to the information to determine the best time to travel, the best route to use and the best mode to choose.

Distributed Memory Environment (DIME)

Collaboration between Prof. Bargiela's Intelligent Simulation Modelling Lab (ISM) and the Nottingham Traffic Control Centre (NTCC) has resulted in the development of a distributed computing system, called DIME, which enables efficient access to real-time traffic data collected by the SCOOT UTC computers. This allows various telematics applications to communicate with each other while maintaining the shared logical view of data. The communication harness is based on a standard TCP/IP protocol and client/server architecture, which makes DIME easily adaptable to various UTC systems and is independent of the physical computing hardware and network type.

There are two components of the DIME:

• the shared memory manager (SMM) task, which owns the shared data buffers,
• the applications program interface DIME-API libraries, which provide a link between user applications and the memory manager.

The SMM component of DIME operates on a closed-loop basis, continually checking the requests to access or to maintain the shared memory data structures. The DIME's API provides facilities for creation and removal of shared memory structures and the read/write access to them. The synchronisation of access to shared memory is implicit in the design of the memory manager task so that the application programs do not need to concern themselves with the complexities of data synchronisation in a distributed computing environment.

The DIME software has been designed to support a wide spectrum of traffic and transportation telematics applications through the provision of customisable static and dynamic data structures and the flexible applications programming interface. The current set of applications that are interfaced to SCOOT via DIME include the following: a predictive macroscopic simulator; a portable traffic/travel information system; a microscopic traffic simulator; and a geographical information system (GIS) based interface to UTC.

The DIME system has been deployed in the Nottingham Traffic Control Centre since 1997 and it has been extensively evaluated using both local- (LAN) and wide-area (WAN) networks. Performance tests have shown that compared to a multiple point-to-point data acquisition, an order-of-magnitude improvement in data throughput is attainable with DIME without affecting the operation of the SCOOT system.

Predictive macroscopic traffic simulator

One of the major criticisms of the current demand-responsive UTC systems is that they react to the already existing traffic situation rather than anticipating the onset of traffic congestion and trying to prevent it. This is primarily because it is both hard to predict the intentions of road users and, more importantly, to assess the accuracy of such predictions so as to ensure that any supervisory traffic control would not be counterproductive.



As a response to this challenge a predictive macroscopic simulator has been developed at the ISM Lab and has been interfaced through DIME to the SCOOT system at the NTCC. The simulator processes the real-time measurements of traffic flows and queue lengths and derives an adaptive model of traffic evolution, which is used to predict not only journey times but also the confidence limits on these predictions i.e. the possible variation of journey times. It is interesting to note that the variation of travel time is considered by most drivers as a more critical piece of information than the travel time itself and appears to be a principal factor in deciding whether to switch to public transport.

Another important advantage of the predictive macrosimulator is that the causal relationship that exists between the accuracy of traffic measurements and the accuracy of traffic predictions is made explicit thus providing a firm basis for making decisions on the expansion and/or improvement of traffic telemetry systems.

Portable traffic/travel information

The idea behind traffic and travel information systems is to assist drivers in selecting among alternative routes according to the current urban traffic situation. Also, very importantly, to facilitate a comparison of the car journey with the alternative modes of transport. Current systems fall into two broad categories: those installed in vehicles and primarily concerned with assisting drivers and those installed on the road-side and intended to assist the non-drivers. Unfortunately, because of this functional split, these systems have a rather limited influence on mode switching decisions.

The portable traffic and travel information system developed at the ISM Lab combines the functionality of both 'in-car' and 'road-side' information systems. The system is implemented using a handheld electronic organiser (PSION) and the GSM telephone. Since both pieces of equipment are standard consumer electronics products, which are not constrained by being part of the car's equipment or being deployed at a particular road-side location, the combined system is expected to have a much broader user base and, by implication, a much greater impact on mode switching decisions.

The integration of the portable information system within the framework of the distributed memory environment (DIME) ensures real-time access to traffic data and the computational results of all distributed applications in the DIME environment. In particular, access to information on confidence limits on journey times is expected to provide a valuable reference for mode switching decisions.

 

Real-time microscopic traffic simulator

Microscopic simulation of urban traffic is a powerful technique that enables the evaluation of the effectiveness of traffic control strategies, the effectiveness of road layout design, the interaction of various traffic parameters, etc. However there are some major limitations associated with this approach such as the need to provide significant computing resources and the need to make various statistical assumptions about individual vehicles.

In order to overcome these limitations a distributed microsimulator (HUTSIM/DIME) has been developed as a result of a research collaboration between the Transportation Research Laboratory at the Helsinki University of Technology and the ISM Lab. HUTSIM/DIME facilitates distributed execution over an arbitrary number of workstations, thus enabling a flexible augmentation of the scale of simulations. Since HUTSIM/DIME processes real-time traffic data, it eliminates the need for various statistical assumptions and produces microsimulation results that reflect much more closely the reality.



By comparing the results of the real-time microsimulations to the measurements of traffic flows and queues one can obtain indirect measurements of prevailing traffic dynamics. These indirect measurements are used to refine traffic control strategies so that they maximise the road capacity while promoting safe driving behaviours.

GIS based interface to UTC

Recognising that traffic systems domain exhibits complex static and dynamic characteristics, a general database model has been developed and implemented using the commercial geographical information system software (Mapinfo). The model allows recording of relationships among entities involved in traffic processes and the representation of causal relationships that are relevant from an application point of view. The integration of varying levels of spatial and temporal detail facilitates derivation of different abstraction levels of data. For example one can interrogate the database about instantaneous traffic flows in a particular link, along a pre-specified route or in a given area. Alternatively one can enquire about averaged flows using an arbitrary averaging period and arbitrary time-shift.

The combination of simulations and the real data are providing a powerful tool for decision making. In this context the capability to abstract and to convey information in an intuitive way through a geographic information system can be of vital importance in analysing wide range of practical environmental issues that enhance the sustainability of cities.

Conclusions

The distributed memory environment DIME has demonstrated its potential for enhancing the value of a UTC/SCOOT system by facilitating retrieval of real-time traffic data without necessitating any software changes to the UTC system and without imposing any significant overheads on the host system. The system facilitated data-sharing among research groups in the UK, Finland and Japan.

The set of distributed applications making use of the real-time traffic data illustrate the point about the added value realised by the DIME system through real-time applications.

The system design is such that any future ITS facility can easily be added to the DIME environment and therefore provide an accessible path forward from Traffic Control System to Demand Management Toolkit.

References

• Bargiela A., Kosonen I, Pursula M, Peytchev E., Granular analysis of traffic data for turning movements estimation, Int. J. of Enterprise Information Systems , vol 2, 2, 2006, 13-27, PDF
• Kosonen I., Bargiela A., Claramunt C., A distributed Traffic Monitoring and Information System, Journal of Geographic Information and Data Analysis , Vol 3, no. 1, June 1999, pp. 30-39, PDF
• Bargiela A., Kosonen I., Pursula M., Peytchev E., Granular analysis of traffic data for turning movements estimation, Proc. 5th Asia Pacific Industrial Engineering and Management Systems Conference, Brisbane, December 2004, PDF
• Kosonen I., Bargiela A., Real-time environment for micro-simulation of urban traffic, Proc. European Simulation Symposium, ESS'2001, Marseille, Oct. 2001, 382-386, PDF
• Hartley J., Bargiela A., Decision support for planning multi-modal urban travel, Proc. European Simulation Symposium, ESS-2001, Marseille, Oct. 2001, 387-391.
• Bargiela A., Peytchev E., Intelligent transportation systems: Towards integrated framework for traffic/transport telematics applications, Proc. IEEE Vehicular Technology Conference, Atlantic City, Oct. 6-10, 2001, PDF
• Penev K, Claramunt K., Bargiela A., A multi-level approach to the integration of traffic data within GIS, Proc. GISUK conference Dec. 2000, PDF
• Kosonen I., Bargiela A., Simulation based traffic information system, Proc. Seventh World Congress on Intelligent Transport Systems, 6-9 November 2000, Turin, Italy, PDF
• Kosonen, I., Bargiela, A., Real time microsimulation of urban traffic, Proc. INFORMS-2000, Salt Lake City, May 2000, PDF
• Peytchev E., Bargiela A., On-line mobile passenger information system, Proc. Int. Workshop on Harbour, Maritime and Industrial Logistics Modelling and Simulation - HMS, Genoa, Sept. 1999, PDF
• Claramunt, C., Bargiela, A. and E. Peytchev, A real-time GIS for the analysis of a traffic system, Proceedings of the 5th IEEE Africon-99 Conference, Cape Town, South Africa, The IEEE Society, Sept. 1999, pp.15-20, PDF
• Bargiela A, Peytchev E., Berry R., Experiences with a distributed traffic telematics environment - portable travel information sys­tem, Proc. IEEE - AFRICON-99 Conf., Sept. 1999, ISBN 0-7803-5546-6, pp.7-12, PDF
• Peytchev E., Bargiela A., Traffic Telematics Software Environment, Proc. European Simulation Symposium ESS-98, Oct. 1998, ISBN 1-56555-147-8, pp.378-382., PDF
• Kosonen I., Bargiela A., Claramunt C., A distributed traffic monitoring and information system, Proc. European Simulation Symposium ESS-98, Oct. 1998, ISBN 1-56555-147-8, pp.355-361, PDF
• Etches, A., Claramunt, C., Bargiela, A., Kosonen, I., An integrated temporal GIS-model for traffic system, Proc. Geographic Information Systems Research Conference (GISUK), April 1998, Edinburgh, UK, PDF
• Peytchev E., Bargiela A., Gessing R., A Predictive Macroscopic City Traffic Flows Simulation Model, Proceedings of European Simulation Symposium ESS-96, Genoa, October 1996, ISBN 1-565555-099-4 (Vol.2), pp.38-42, PDF
• Argile A., Peytchev E., Bargiela A., Kosonen I., DIME: A Shared Memory Environment for Distributed Simulation, Monitoring and Control of Urban Traffic, Proceedings of European Simulation Symposium ESS'96, Genoa, 1996, pp.152-156, PDF
• Peytchev E., Bargiela A., Parallel simulation of city traffic using PADSIM, Proc. of Modelling and Simulation Con­ference ESM-95, Prague, June 1995, ISBN 1-56555-080-3, pp. 330-334, , PDF
• Peytchev E., Bargiela A., Micro simulation of city traffic in support of predictive operational control, Proc. 10th Int. Conf. on Systems Engineering, ISCE-94, Coventry, Sept. 1994, ISBN 0 905 94923 4, pp. 934-941
• Argile A., Bargiela A., Using X11 Windows to provide shared task memory in distributed computer systems, , Int. Conf. On Computing and Control in Water Industry, Leicester, 1993.