Development of a unified framework for optimal control and estimation of nonlinear stochastic systems based on path integral analysis

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K18089
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Control engineering/System engineering
• Research Institution: Osaka University (2017); Hiroshima University (2015-2016)
• Principal Investigator: Satoh Satoshi 大阪大学, 工学研究科, 講師 (60533643)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: 確率制御 / 非線形制御 / 最適制御 / 小型ヘリコプタ

Outline of Final Research Achievements

We have proposed an iterative solution method for nonlinear stochastic optimal control problems using stochastic analysis and statistical sampling. In this project, first, we have extended this method to a nonlinear filtering problem. Second, a modeling method for a nonlinear stochastic system has been investigated, which is based on the parameter identification with the nonlinear filtering method. Third, we have provided a unified framework for stochastic optimal control, state estimation and stochastic system modeling by the path integral analysis approach. Finally, we have performed experimental verification of the proposed method with small unmanned helicopters. Optimal flight trajectories for the helicopters taking probabilistic uncertainties into account have been obtained.

Free Research Field

制御工学