← Back to list of papers of the 2022 EuroGNC conference |
CEAS EuroGNC 2022 |
||||||
Data-efficient capture region estimation for tactical missile using active sampling based Gaussian process classification |
||||||
|
||||||
Abstract: An efficient capture region estimation algorithm is proposed for a missile. The Gaussian process classifier and active sampling technique are combined to improve the accuracy of the predictive model while efficiently using data. The key idea is to actively sample the data required for the learning method, which uses the information on the uncertainty predicted from the Gaussian process classifier. Data necessary for training are obtained using a planar missile simulator, and impact-angle control composite control guidance is used for intercepting the target while satisfying the impact angle requirement. Through numerical simulation, it is shown that the accuracy of the trained model gradually increases during training progress, and the performance of the finally-trained predictive model is verified. It is also shown that the efficiency of the data usage and the accuracy of the trained model are relatively better than a Gaussian process classifier trained without active sampling. | ||||||
Keywords: Gaussian Process Classification; Active Sampling; Capture Region Estimation; Impact Angle Control Composite Guidance | ||||||
View PDF CEAS-GNC-2022-072 |
Youngjun Lee, Sangmin Lee, Youdan Kim: Data-efficient capture region estimation for tactical missile using active sampling based Gaussian process classification. Proceedings of the 2022 CEAS EuroGNC conference. Berlin, Germany. May 2022. CEAS-GNC-2022-072. |
BibTeX entry: @Incollection{CEAS-GNC-2022-072, authors = {Lee, Youngjun and Lee, Sangmin and Kim, Youdan}, title = {Data-efficient capture region estimation for tactical missile using active sampling based Gaussian process classification}, booktitle = {Proceedings of the 2022 {CEAS EuroGNC} conference}, address = {Berlin, Germany}, month = may, year = {2022}, note = {CEAS-GNC-2022-072} } |