SISG4HEI_Alpha: Alpha version of simulated indoor scenario generator for houses with elderly individuals

Can Jiang, Akira Mita

Research output: Contribution to journalArticlepeer-review

Abstract

Developers improve and test the sensor networks and robots of smart houses by conducting simulations. However, they need to build scenarios in the simulators manually, which takes a lot of time. Manually built scenarios are also few in number. Moreover, developers have to repeat this procedure whenever they conduct a simulation with the same scenario on another simulator. To address these issues, we have developed a simulation tool (alpha version), called SISG4HEI_Alpha, which produces diverse simulated indoor scenarios of houses with elderly individuals for multiple simulators. It consists of a GUI, database, three generators, and interfaces. The spatial attribute generator produces virtual spatial data necessary for building the models of the house and furniture. The activity schedule and travel pattern generator generate resident's information to control the virtual agent within a simulator. One interface converts the generated data into simulated PIR sensor records, which are simulation results of the OPENSHS simulator. The two other interfaces convert data into scenarios of the Stage and Gazebo simulators. SISG4HEI_Alpha is modular and extensible; it is easy to improve the generators and/or add new interfaces.

Original languageEnglish
Article number101963
JournalJournal of Building Engineering
Volume35
DOIs
Publication statusPublished - 2021 Mar

Keywords

  • Automatic simulation
  • Elderly people living alone
  • Machine learning datasets
  • Simulation scenario generation
  • Smart house simulator

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Architecture
  • Building and Construction
  • Safety, Risk, Reliability and Quality
  • Mechanics of Materials

Fingerprint Dive into the research topics of 'SISG4HEI_Alpha: Alpha version of simulated indoor scenario generator for houses with elderly individuals'. Together they form a unique fingerprint.

Cite this