Simulation of Static and Dynamic Evacuation Algorithms in Intelligent Buildings Using Unity3D

Abstract

Dynamic escape route guidance in intelligent buildings offers a great opportunity to evacuate buildings more efficiently and safely in comparison to static escape route guidance. A disadvantage of static escape route guidance is that the system is not able to react to temporary events in the building. In a worst-case scenario, the escape route leads directly into the fire or hazard zone. To investigate this risk, two algorithms for building evacuation have been developed. The first algorithm (static), demonstrates the static evacuation from the building. This is designed to demonstrate the behavior based on rescue and escape route planning, which is currently common used in existing and new buildings. The second algorithm (dynamic) shows how evacuation could take place in an intelligent building using an autonomous system. This implies that the evacuation system is able to react dynamically to the hazards (e.g. fire, toxic gases, or blocked corridors) and replan the escape routes in realtime and visualize them to the occupants inside the building. Both algorithms were investigated on the parameters: Escape route blocked, evacuation time, and length of escape path. The comparison of the algorithms was investigated for randomly generated scenarios and scenarios based on domain knowledge. This study has shown that dynamic escape route guidance has a high potential for improving building evacuation in an emergency situation.