A project for "Automatically Generating Advisory Text for Road Maintenance Vehicle Routing" jointly funded by EPSRC and Aerospace & Marine International

During the winter months local Scottish councils have the difficult task of keeping both minor and major roads in their region safe and clear of snow. Making effective use of road maintenance vehicles relies upon their efficient routing and the correct application of salt and grit to the roads. Road maintenance vehicle routing is based on detailed weather and road surface temperature predictions generated by complex weather computer simulation models.

RoadSafe is a collaborative project between the Computing Science Department at Aberdeen University and Aerospace & Marine International. The RoadSafe project aims to build upon the expertise Aerospace & Marine International has in weather forecasting and the expertise the Computing Science Department at Aberdeen University has in building real world Natural Language Generation Systems.

The RoadSafe project will:-

• use Knowledge Aquisition techniques to understand how humans write textual instructions for road maintenance vehicle routing.
• produce a system capable of automatically evaluating a region's geographical data combined with the weather forecast for 10'000s of points in that region to provide textual routing and de-icer spread rate instructions.
• utilise Aerospace & Marine International's expert forecasters in order to post-edit generated advisory texts and therefore improve the performance of the system.

The main objective of the project is to use the advisory texts produced by RoadSafe as a guide to local councils for grit and salting applications during the winter.

• Dr. Ehud Reiter
• Dr. Yaji Sripada
• Ross Turner
• Ian Davy (Aerospace & Marine International)

• Turner R., Sripada S., Reiter E. Generating Approximate Geographic Descriptions. To appear in proceedings of ENLG-2009, Athens, Greece, 30-31 March 2009. .(PDF)
• Turner R., Sripada S., Reiter E. and Davy I. (2008). Using Spatial Reference Frames to Generate Grounded Textual Summaries of Georeferenced Data In Proceedings of INLG08, Salt Fork, Ohio, USA, 12-14th June 2008 .(PDF)
• Turner R., Sripada S., Reiter E. and Davy I. (2008). Building a Parallel Spatio-Temporal Data-Text Corpus for Summary Generation. To Appear in Proceedings of the LREC2008 Workshop on Methodologies and Resources for Processing Spatial Language, Marrakech, Morocco, 31 May 2008 .(PDF)
• Turner R., Sripada S., Reiter E. and Davy I. (2007). Selecting the Content of Textual Descriptions of Geographically Located Events in Spatio-Temporal Weather Data. In Applications and Innovations in Intelligent Systems XV,pages 75-88 .(PDF)
• Turner R., Sripada S., Reiter E. and Davy I. (2006). Generating Spatio-Temporal Descriptions in Pollen Forecasts. Proceedings of EACL06 Companion Volume.(PDF)

• System 'to help winter crashes'
• Press release

Related Projects

• How Was School Today?
• Atlas.txt
• BabyTalk
• SumTime
• ScubaText

Collaborators

• Aerospace and Marine International

Demos

• A simple demonstration of spatial data analysis and automatic generation of pollen forecasts for Scotland.
• Simple computer generated spatial descriptions for single point locations

---

For more information contact Ross Turner.