Validation of Automotive Control Systems Using Driving Data from Real World

実走行データを活用した制御システム自動検証手法の開発

Verification and validation of large-scale and complicated automotive control systems remains a major challenge. Regarding this challenge, the authors have developed search based testing (SBT), that is the verification method to search for input scenarios that lead to undesirable behavior. Although SBT is useful as a verification method applicable to large-scale and complicated systems, input scenarios not suitable for actual development process are often detected, and it causes an increase in work man-hours. To realize the efficient search, the authors developed a novel SBT with input scenario evaluator based on the driving conditions distribution obtained from the real driving data.大規模・複雑化する制御ソフトの品質確保は重要課題である。これに対し、望ましくない振る舞いにつながる入力シナリオを自動探索するサーチベーステストを開発してきた。本研究では探索効率の向上を目指し、実走行データより得られる実走行条件分布を基にした新たな入力シナリオ生成機能を開発した。

Language

  • English
  • Japanese

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01660604
  • Record Type: Publication
  • Source Agency: Japan Science and Technology Agency (JST)
  • Files: TRIS, JSTAGE
  • Created Date: Jan 31 2018 3:23PM