2017 Fiscal Year Research-status Report
Robust and Adaptive Distributed Control of Discrete-Event Systems: A Supervisor Localization Based Approach
| Project/Area Number |
16K18122
|
|---|
| Research Institution | Osaka City University |
|---|
Principal Investigator |
蔡 凱 大阪市立大学, 大学院工学研究科, 准教授 (20728266)
|
|---|
| Project Period (FY) |
2016-04-01 – 2020-03-31
|
| Keywords | 離散事象システム / 分散制御 / オートマトン |
|---|
| Outline of Annual Research Achievements |
In the second year, the following new results have been achieved.
1. Theoretical results: We further developed partial-observation distributed control with the new concept of relative observability/co-observability. In this setting, some events of the system are unobservable due to lack of sensors in practice.
2. Experiment results: We tested distributed controllers obtained by supervisor localization on the experiment system we developed in the first year. The results are summarized, and will serve as the basis for comparison with more practical distributed controllers to be developed later.
|
| Current Status of Research Progress |
Current Status of Research Progress
2: Research has progressed on the whole more than it was originally planned.
Reason
In the second year (2017.04-2018.03), we have achieved 5 journal papers publications, 1 conference paper publication. In addition to theoretical development, we have also tested distributed control synthesized by supervisor localization on the experiment testbed that we built in the first year.
|
| Strategy for Future Research Activity |
We are going to develop adaptive distributed controllers with learning capability, such that these controllers are able to adapt to changes of the environment as well as deal with large state space. We will continue to develop scalable distributed control, whose control strategies do not depend on the number of agents. In addition, we will test the developed adaptive and scalable distributed controllers on our experiment testbed by using real mobile robots, with special focus on comparing with the experiment data we collected this year.
|
| Causes of Carryover |
In the next fiscal year, I will host Prof. Murray Wonham in Osaka City University for research collaboration, which is important for progressing my Kakenhi project. This was not part of the plan, and therefore certain budget is saved to be used for this purpose.
|
