Survivability based Optimal Air Combat Mission Planning with Reinforcement Learning

Başpınar B., Koyuncu E.

IEEE Conference on Control Technology and Applications (CCTA), Copenhagen, Danimarka, 21 - 24 Ağustos 2018, ss.664-669 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Basıldığı Şehir: Copenhagen
  • Basıldığı Ülke: Danimarka
  • Sayfa Sayıları: ss.664-669
  • İstanbul Teknik Üniversitesi Adresli: Evet

Özet

The mission-threat analysis is the key factor in air combat missions. The strategic success evaluation of a mission is done through survivability assessment. The mission assessment process begins with the modeling of threats in which a hostile environment can be constructed through these models. Then, based on the environment model including stochasticity, an optimum strategy maximizing the aircraft's survivability is obtained. This study proposes a survivability analysis and optimal mission planning methodology using reinforcement learning for an aircraft flying in a human-made and hostile natural environment.