2022 Fiscal Year Research-status Report
A novel Multi-Agent Model with Adversarial Mobility Learning for Epidemic Simulation at the Community Level
| Project/Area Number |
22K11918
|
|---|
| Research Institution | University of Tsukuba |
|---|
Principal Investigator |
Aranha Claus 筑波大学, システム情報系, 助教 (80629858)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
BOGDANOVA ANNA 筑波大学, システム情報系, 助教 (70924463)
長谷部 浩二 筑波大学, システム情報系, 准教授 (80470045)
|
|---|
| Project Period (FY) |
2022-04-01 – 2025-03-31
|
| Keywords | simulator / multi agent systems / machine learning / urban planning / natural disasters / evacuation / city generation |
|---|
| Outline of Annual Research Achievements |
In this year, first we performed a survey on agent-based models of disease spread simulator, understanding the different approaches for the simulation of COVID-19 disease around the world.
Based on this survey, we have designed and implemented a first version of the Community Level Epidemic Simulator (CES), and named it Koudou. This simulator reproduces the campus of the University of Tsukuba and its neighborhood. It represents agents as workers and students who go around the city for their day-to-day activities. The simulation tracks the spread of an airborne disease like COVID-19 indoors and outdoors, and takes into account variation on mask usage. The simulation also reproduces the evacuation from a large earthquake, and measures the differences in disease spread based on gatherings in evacuation centers.
We have also developed an "Intelligent City Planning Model (ICPM)". The ICPM uses uses Machine Learning to understand and integrate demographic and geographical data, Land Use and Transport (LUTI), to generate a virtual city. We demonstrated that the ICPM can learn from real world urban data and re-create realistic designs of cities based on the needs of simulated agents. This is a first step for learning advanced mobility models for agents in simulation.
Both works above were accepted for presentation at the International ALIFE 2023 conference, and have been made available to the public as open-source projects.
|
| Current Status of Research Progress |
Current Status of Research Progress
3: Progress in research has been slightly delayed.
Reason
- The development of the Community Epidemic Simulator (CES) was "Step 2" of the proposed research plan. It has proceeded smoothly, according to plan. The simulator has all the necessary functionality for simulating disease spread and the effects of disaster evacuation on the spread of disease. However, it still needs further development so that it is easy to use for the end-user. This will be the future work for this step.
- The development of the "Intelligent Mobility Model" (IMM) was "Step 1" of the proposed research plan. It is slightly delayed. This step depends on the availability of fine-grained mobility data in Japan. However we were not able to secure this data last year. We have found some data from mobile companies, but the resolution of that data was too low (500 meters and above), and could not be used for adversarial learning. We are considering alternative sources of data, such as WIFI records.
- The development of the Intelligent City Planning Model (ICPM) is an alternative idea to develop a new mobility model without using fine-grained user data. The hypothesis is that mobility is closely linked to city development, and we will be able to use this model to develop better mobility patterns without using high resolution data. We had good progress on this approach this year.
|
| Strategy for Future Research Activity |
- We will continue the development of the Community Evacuation Simulator, with two focuses: (1) The ability of users to add plug-in model extensions, such as agent communication, agent psychology, and public transportation, (2) Increased visualization ability of the results, such as maps, animations and charts. These two developments are in preparation to the validation by the community stage, which is predicted to happen in 2024.
- We will advance the development of the Intelligent City Planning Model as a replacement of IMM model, so that agent mobility can be learned together with the development model of the city itself. Using the hypothesis that the development of the city and its use and mobility are intrinsically linked.
- We will search alternative sources of data to satisfy the requirements of original IMM model plan, in parallel with the ICPM model research.
- As a preparation to the validation stage, we have begun collaborative work with a researcher on disaster resilience and earthquake evacuation from the University of Kobe. We are planning a join paper this year that uses the CES model together with actual evacuation training data to validate the current mobility aspects of the model and determine the best areas for improvement.
|
