Models@run.time Framework for Graceful Degradation

2022 Fiscal Year Final Research Report

• Project/Area Number: 18H03225
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Single-year Grants
• Section: 一般
• Review Section: Basic Section 60050:Software-related
• Research Institution: Waseda University
• Principal Investigator: TEI Kenji 早稲田大学, 理工学術院, 准教授(任期付) (40434295)
• Co-Investigator(Kenkyū-buntansha): 本位田 真一 早稲田大学, 理工学術院, 教授(任期付) (70332153)
• Project Period (FY): 2018-04-01 – 2022-03-31
• Keywords: 自己適応システム / Models@run.time / 離散制御器合成 / モデル学習

Outline of Final Research Achievements

We aimed to realize Graceful Degradation, which guarantees maximum safety even in the case of "changes that were not assumed at the time of development". For this purpose, this research established Models@run.time techniques in which the system itself utilizes the model at runtime to realize self-adaptation with guarantees. Specifically, we have established techniques that (1) reflect changes that were not assumed during development in the model and (2) automatically synthesize at runtime a behavior specification that guarantees safety under the updated environmental model. We also developed a models@run.time framework that reflects the established technology, and clarified the its effectiveness and limitations through evaluation experiments on IoT systems and robot systems.

Free Research Field

ソフトウェア工学

Academic Significance and Societal Importance of the Research Achievements

開発時の想定のみに頼る従来の安全性保証技術では，本質的に想定漏れを避けることが困難な近年のソフトウェアシステムで十分な安全性を保証することができない．近年のIoTシステムやCPSが対象とするオープン環境ではシステムの動作に影響を与えうる事象が無数に存在する．あらゆる可能性を想定しようとすると工数が増大し，また想定漏れは本質的に防ぎきれない．そこで本研究では開発時の想定に漏れた環境変化が起きてもシステムが即応的に適応し，その時点で可能な最大限の安全性を保証するよう段階的に動作を変更するGraceful Degradationを実現する技術を構築した．