Human–Behaviour Simulation

Human movement and behaviour simulation have a long and rich history in the literature. The simulation of human dynamics, from person-to-person interactions to global-scale transportation networks, affords a plethora of revolutionary predictive and analytical powers across several fields. We make use of  biomechanical modelling techniques similar to bipedal locomotion of humans to achieve realistic simulations.

Modeling Human–Building Interactions

The maturity in modeling human movements and behaviours facilitates the investigation of how the design of semantically meaningful environments impacts the movement and occupancy of its inhabitants, and is a precursor to informing the design of buildings with respect to human-centric performance criteria. To this end, our research outlines:

  • Wayfinding in Complex Environments

  • Integrated Modeling of Individuals, Crowds, and Building Semantics

  • Joint Modeling of Human-Building Interactions

  • Machine Learning for Human-Building Interaction

Environment Design Analytics and Optimization

During an environment design process, architects and game-level designers have to examine a vast set of solutions to identify efficient parameters which would increase the design performance by following given set of constraints. However, examining each solution iteratively is time consuming, and perhaps less efficient. In our research, we develop computer-aided design (CAD) tools which allow predictive analysis and evaluation of building designs using human–behaviour simulations.

Cross-Platform Human-Simulation Service

Running human behaviour simulations in real-scale complex environments present high integration costs into building design workflows as they are often decoupled from architectural modeling tools. They require significant infrastructure and domain expertise, hindering the designers' abilities to seamlessly simulate, analyze, and incorporate dynamics of human-building interactions into their preferred design workflows. In our research, we develop a democratized approach using an on-demand service-based workflow for human behaviour simulation and analysis of human-building interactions.