Symmetric models for radar response modeling

Colin Kohler, Nathan Vaska, Ramya Muthukrishnan, Whangbong Choi, Jung Yeon Park, Justin Goodwin, Rajmonda Caceres, Robin Walters
Proceedings of the 2nd NeurIPS Workshop on Symmetry and Geometry in Neural Representations, PMLR 228:303-323, 2024.

Abstract

Many radar applications require complex radar signature models that incorporate characteristics of an object’s shape and dynamics as well as sensing effects. Even though high-fidelity, first-principles radar simulators are available, they tend to be resource-intensive and do not easily support the requirements of agile and large-scale AI development and evaluation frameworks. Deep learning represents an attractive alternative to these numerical methods, but can have large data requirements and limited generalization ability. In this work, we present the Radar Equivariant Model (REM), the first SO(3)-equivaraint model for predicting radar responses from object meshes. By constraining our model to the symmetries inherent to radar sensing, REM is able to achieve a high level reconstruction of signals generated by a first-principles radar model and shows improved performance and sample efficiency over other encoder-decoder models.

Cite this Paper

BibTeX
@InProceedings{pmlr-v228-kohler24a, title = {Symmetric models for radar response modeling}, author = {Kohler, Colin and Vaska, Nathan and Muthukrishnan, Ramya and Choi, Whangbong and Park, Jung Yeon and Goodwin, Justin and Caceres, Rajmonda and Walters, Robin}, booktitle = {Proceedings of the 2nd NeurIPS Workshop on Symmetry and Geometry in Neural Representations}, pages = {303--323}, year = {2024}, editor = {Sanborn, Sophia and Shewmake, Christian and Azeglio, Simone and Miolane, Nina}, volume = {228}, series = {Proceedings of Machine Learning Research}, month = {16 Dec}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v228/main/assets/kohler24a/kohler24a.pdf}, url = {https://proceedings.mlr.press/v228/kohler24a.html}, abstract = {Many radar applications require complex radar signature models that incorporate characteristics of an object’s shape and dynamics as well as sensing effects. Even though high-fidelity, first-principles radar simulators are available, they tend to be resource-intensive and do not easily support the requirements of agile and large-scale AI development and evaluation frameworks. Deep learning represents an attractive alternative to these numerical methods, but can have large data requirements and limited generalization ability. In this work, we present the Radar Equivariant Model (REM), the first SO(3)-equivaraint model for predicting radar responses from object meshes. By constraining our model to the symmetries inherent to radar sensing, REM is able to achieve a high level reconstruction of signals generated by a first-principles radar model and shows improved performance and sample efficiency over other encoder-decoder models.} }
Endnote
%0 Conference Paper %T Symmetric models for radar response modeling %A Colin Kohler %A Nathan Vaska %A Ramya Muthukrishnan %A Whangbong Choi %A Jung Yeon Park %A Justin Goodwin %A Rajmonda Caceres %A Robin Walters %B Proceedings of the 2nd NeurIPS Workshop on Symmetry and Geometry in Neural Representations %C Proceedings of Machine Learning Research %D 2024 %E Sophia Sanborn %E Christian Shewmake %E Simone Azeglio %E Nina Miolane %F pmlr-v228-kohler24a %I PMLR %P 303--323 %U https://proceedings.mlr.press/v228/kohler24a.html %V 228 %X Many radar applications require complex radar signature models that incorporate characteristics of an object’s shape and dynamics as well as sensing effects. Even though high-fidelity, first-principles radar simulators are available, they tend to be resource-intensive and do not easily support the requirements of agile and large-scale AI development and evaluation frameworks. Deep learning represents an attractive alternative to these numerical methods, but can have large data requirements and limited generalization ability. In this work, we present the Radar Equivariant Model (REM), the first SO(3)-equivaraint model for predicting radar responses from object meshes. By constraining our model to the symmetries inherent to radar sensing, REM is able to achieve a high level reconstruction of signals generated by a first-principles radar model and shows improved performance and sample efficiency over other encoder-decoder models.
APA
Kohler, C., Vaska, N., Muthukrishnan, R., Choi, W., Park, J.Y., Goodwin, J., Caceres, R. & Walters, R.. (2024). Symmetric models for radar response modeling. Proceedings of the 2nd NeurIPS Workshop on Symmetry and Geometry in Neural Representations, in Proceedings of Machine Learning Research 228:303-323 Available from https://proceedings.mlr.press/v228/kohler24a.html.

Related Material