Developing software to aid disaster management

Web and Internet Science (WAIS) researchers from the University of Southampton have been involved in developing essential provenance software for the award-winning ORCHID project.

ORCHID is a £10m major research project tackling a key challenge in modern computer science – how humans and intelligent software systems can seamlessly and effectively work together to aid the management of a disaster environment such as the Haiti earthquake and the Fukushima nuclear tragedy.

The project explored the use of unmanned autonomous vehicles (UAVs) fitted with audio-visual equipment and crowd-sourced information to improve the response of emergency services and populations to disaster management.

The human-agent collective technology has been used in the aftermath of the Nepal earthquake to help save lives. Crowd-sourced data enabled rescuers to map settlements across the affected areas and identify aid, and search and rescue priorities. Crowd reports are collected and verified by UAVs, first responders are deployed and assignments may be changed depending on new information gathered.

However, the amount of data being generated during an emergency situation means the information is continually changing and needs to be constantly updated.

ORCHID co-investigator Dr Sarvapali Ramchurn, from Southampton’s Agents, Interaction and Complexity Research Group, said: “In order to ensure everybody involved gets the most up-to-date information to get the job done, we combine human and machine intelligence to optimise the collection, analysis and filtering of data coming from the disaster zone.â€?

His Southampton colleague Dr Dong Huynh, a WAIS researcher in the fields of trust, reputation and provenance, led a team developing software that would ensure the information available to responders was as accurate and up-to-date as possible.

He said: “When new information comes into the system this may invalidate an existing assignment. We have developed provenance management software that recognises these changes and feeds them back to the task allocation commander, thereby providing a continuous feedback loop of information.â€?

This was made possible by tracking and reporting the provenance of every piece of data or decision generated in the whole disaster management system. To enable them to effectively implement this, the team developed original software and tools to manage provenance using the latest PROV standards by the World Wide Web Consortium – the standardisation body for Web standards.

Professor Luc Moreau, ORCHID co-investigator and co-chair of the provenance standardisation group at the World Wide Web Consortium, said: “The provenance standard PROV was critical in describing changes to our knowledge about the situation on the ground, identifying all the planned activities that were dependent on that revised knowledge, and triggering re-planning where appropriate.â€?

ORCHID recently took top prize in the data and connectivity category at the
Collaborate to Innovate Awards organised by The Engineer – the UK’s longest-running engineering publication. The awards were aimed at identifying some of the UK’s most impressive and innovative examples of engineering collaboration.