Dr Victoria Hodge

Back to the Advanced Computer Architectures page

Researcher Profile: Dr Victoria Hodge

FREEFLOW: Intelligent Decision Support for Traffic Management

The objective of Freeflow is: to improve traffic network management and operation by turning data into intelligence.

About my research

Our research is in the general areas of Neural Networks, Pattern Recognition and Data Mining in the Advanced Computer Architectures Research Group. More specifically, we are involved in developing a system that will detect patterns and anomalies in traffic flows, use pattern matching to find similar historical patterns. We then take suitable measures to improve the current traffic flow using both historical knowledge and traffic modelling.

Traffic at night (Fig1. Traffic generates data) & Neural network (Fig2. Data are processed in a neural network)


The work forms part of the Freeflow project involving 15 partners from academia, industry and government. The objective of Freeflow is: to improve traffic network management and operation by turning data into intelligence. Currently more and more detailed and timely data about transport networks are being collected. In FREEFLOW, these data will be integrated and mined to actively manage traffic.

Freeflow will operate on three sites:

  • In London, Freeflow will analyse Park Lane and Hyde Park Corner, a key London route which is as busy as many UK motorways, and select appropriate traffic signal settings and other measures to mitigate congestion.
  • In Maidstone, the project focuses on managing the urban / inter-urban interface and displaying sets of messages on variable message signs to improve travel times and reduce congestion.
  • In York, work will focus on bus punctuality and selecting the most appropriate traffic signal plans and messages to display to maintain punctuality.

York FREEFLOW work

At University of York, we are developing a k-nearest neighbour based pattern matching tool [1, 2] using the Advanced Uncertain Reasoning Architecture (AURA).  AURA is based on Correlation Matrix Memories (CMMs) which are binary associative neural networks.  CMMs can store large amounts of data and allow fast searches.  We convert traffic data variables (such as data from sensors embedded in the road or from buses) into vectors using a quantisation process. These vectors are then stored in a historical database of vectors in the CMM.

As new traffic data is generated, we turn this new data into a query vector using the quantisation process.  This vector is applied to AURA to find the k best matching historical time periods through vector similarity using kernels and incorporating spatio-temporal aspects. We finally provide advice to the traffic operator by cross-referencing operator logs for traffic control interventions made during the k best matching time periods; calculate a quality score for each of these interventions (how well it worked); and, thus, recommend to the operator the intervention likely to be most effective for the current situation.

A Google map with traffic control advise data (Fig3. Traffic control advice is generated)


Figure 3. Traffic control advice is generated.

[1] Victoria J. Hodge, Rajesh Krishnan, Jim Austin & John Polak. A Computationally Efficient Method for Online Identification of Traffic Incidents and Network Equipment Failures. Presented at, Transport Science and Technology Congress: TRANSTEC 2010, Delhi, April 4-7, 2010

[2] Rajesh Krishnan, Victoria J. Hodge, Jim Austin & John Polak. A Computationally Efficient Method for Online Identification of Traffic Control Intervention Measures. Presented at, 42nd Annual UTSG Conference, Centre for Sustainable Transport, University of Plymouth, UK: January 5-7, 2010.

Want to know more?

Visit Victoria's Personal Page or our Advanced Computer Architectures research group web pages.

Back to Top