Incorporating real-time data into agent-based crowd simulations using dynamic data assimilation

 

Project overview 

We are building an agent-based model of people’s movements through the Connected Places Smart Building in London, which is equipped with sensors. The model will incorporate real-time sensor data by establishing a method of dynamic data assimilation. This is not currently standard practice for crowd simulations. The project deals with preparing the sensor data for use in the model. 

 

Data and methods  

The source of the data are the sensors in the Smart Building in London. This building is a sensor testbed maintained by Connected Places Catapult. The sensors measure parameters such as CO2 levels, light levels, and noise levels which reflect the internal environment of the building. The data are available through an API. To build the model, the data must be accessible programmatically. This requires a means to access the API, a database to store the data, and a means to retrieve the data in order to plug it into the model. 

 

Key findings  

The primary output of the project was a series of scripts written in Python and SQL. ‘scraper.py’ is a Python module for logging into and accessing the Smart Building API, and to retrieve data for storage in variables. It is also possible to plot figures from any sensor (Figure 1). This was accompanied by scraperplot.py which is a script which uses ‘scraper.py’ to plot through options and input in the command line. 

 

Scripts for creating an SQL database and storing data from the sensors collected using ‘scraper.py’ were created: ‘create_database.sql’ and ‘database.py’ (Figure 2). Lastly, ‘databaseplot.py’ is a user-friendly script which you can use to plot specific sensors and parameters from a specific time point, as well as aggregating sensor readings from a single room (Figure 3). The scripts, and documentation and notebooks for use, are available on github: 

https://github.com/TCR1990/SmartBuilding-master  

Also as part of this project, we conducted a review of software for agent-based modelling of crowd simulations: 

https://urban-analytics.github.io/dust/docs/ped_sim_review.pdf  

 

Value of the research  

This project forms part of the wider Data Assimilation for Agent-based Models (DUST) project, and has real-world implications for improving the efficiency of management of events within buildings. 

 

Insights 

  • These are the first steps towards building an agent-based model of the Connected Places Smart Building 
  • The scripts generated enable data retrieval, storage, and plotting from the Smart Building API 

 

People 

Tom Richards, LIDA Intern

Nicholas Malleson, Professor of Geography, LIDA, University of Leeds 

Jonathon Ward, Lecturer in Mathematics, LIDA, University of Leeds 

Minh KieuLecturerThe University of Auckland 

Tamar Loach, Data Science Team LeadCatapult 

 

Partners 

Connected Places Catapult 

  

Funders 

Part of the DUST project, funded by the European Research Council