Government-backed innovation agency, the Transport Systems Catapult (TSC), and Sheffield University have built a computer model to help London Underground understand how changes to stations will impact passengers' movements and safety.
The prototype uses a simulation of Canary Wharf Station, through which the system can send large numbers of virtual travellers exhibiting typical behaviour patterns – eg moving straight into or out of the station, stopping to buy tickets or check directions on wall maps, or diverting towards shops.
Modelling a herd of cats? Software mimics typical behaviour
Sheffield University is using its FLAME (Flexible Large-scale Agent Modelling Environment) Graphics Processing Unit technology to generate the people flows.
Researchers can then mingle with the resulting 'crowds', wearing a Swedish-made virtual reality headset - originally developed for gaming and now being deployed for the first time in the UK - and walking on an omnidirectional 'treadmill'-style rolling belt set in the TSC Visualisation Laboratory's floor.
The system accurately tracks their movements, changes of direction, avoidance of 'collisions' and walking speeds of up to 07km/h for detailed analysis.
The TSC sees the system saving transport operators the costs of real-world trials and badly thought-out changes which could, for example, adversely affect passengers' escape routes after an incident.