Pose graph optimization consists in the estimation of a set of poses from pairwise measurements and provides a formalization for many problems arising in mobile robotics and geometric computer vision. In this letter, we consider two-dimensional pose estimation problems in which a subset of the measurements is spurious. Our first contribution is to develop robust estimators that can cope with heavy-tailed measurement noise, hence increasing robustness to the presence of outliers. Since the resulting estimators require solving nonconvex optimization problems, we further develop convex relaxations that approximately solve those problems via semidefinite programming. We then provide conditions for testing the exactness of the proposed relaxations. Contrary to existing approaches, our convex relaxations do not rely on the availability of an initial guess for the unknown poses, hence they are more suitable for setups in which such guess is not available (e.g., multirobot localization, recovery after localization failure). We tested the proposed techniques in extensive simulations, and we show that some of the proposed relaxations are indeed tight (i.e., they solve the original nonconvex problem exactly) and ensure accurate estimation in the face of a large number of outliers.
Convex Relaxations for Pose Graph Optimization With Outliers / Carlone, Luca; Calafiore, GIUSEPPE CARLO. - In: IEEE ROBOTICS AND AUTOMATION LETTERS. - ISSN 2377-3766. - STAMPA. - 3:2(2018), pp. 1160-1167. [10.1109/LRA.2018.2793352]
Convex Relaxations for Pose Graph Optimization With Outliers
Giuseppe Calafiore
2018
Abstract
Pose graph optimization consists in the estimation of a set of poses from pairwise measurements and provides a formalization for many problems arising in mobile robotics and geometric computer vision. In this letter, we consider two-dimensional pose estimation problems in which a subset of the measurements is spurious. Our first contribution is to develop robust estimators that can cope with heavy-tailed measurement noise, hence increasing robustness to the presence of outliers. Since the resulting estimators require solving nonconvex optimization problems, we further develop convex relaxations that approximately solve those problems via semidefinite programming. We then provide conditions for testing the exactness of the proposed relaxations. Contrary to existing approaches, our convex relaxations do not rely on the availability of an initial guess for the unknown poses, hence they are more suitable for setups in which such guess is not available (e.g., multirobot localization, recovery after localization failure). We tested the proposed techniques in extensive simulations, and we show that some of the proposed relaxations are indeed tight (i.e., they solve the original nonconvex problem exactly) and ensure accurate estimation in the face of a large number of outliers.File | Dimensione | Formato | |
---|---|---|---|
08258864.pdf
non disponibili
Tipologia:
2a Post-print versione editoriale / Version of Record
Licenza:
Non Pubblico - Accesso privato/ristretto
Dimensione
465.54 kB
Formato
Adobe PDF
|
465.54 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11583/2715547
Attenzione
Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo