Key technology of lightweight Web3D online planning of metro fire escape

Abstract

Virtual training of metro fire evacuation definitely becomes considerably more efficient and convenient immigration of trainees from the PC desktop to mobile web browsers. However, web browsers face extreme difficulty in sustaining online visualization of large-scale building information modeling (BIM) data on metro, FDS smoking data, crowding passengers, and planning of fire evacuation path simultaneously due to the weak rendering capability and limited networking bandwidth. Herein, a novel lightweight Web3D technology roadmap is proposed to address these problems. First, the lightweight preprocessing of metro BIM data is performed by semantic and geometric retrieval of repetitive entities. Second, crowding passengers are visualized online by reusing several avatars and behavior randomly in terms of textures and scales in a lightweight manner. Third, substantial smoking data are reduced by redundancy elimination and voxel normalization and rendered online with spirit texture particle. Finally, the backbone escape paths are extracted by clustering the virtual evacuation traces that are collected by virtual reality devices, such as HTC VIVE, as initial heuristics, and a lightweight metro fire evacuation planning evacuation based on adaptive ant colony optimization is realized on mobile web browsers. Numerous tests have been conducted online, and experimental results show that the proposed method can be loaded remotely in a few seconds and visualized on mobile web browsers smoothly.