-
A scattering theory construction of dynamical solitons in 3d
Authors:
Istvan Kadar
Abstract:
We study the energy critical wave equation in 3 dimensions around a single soliton. We obtain energy boundedness (modulo unstable modes) for the linearised problem. We use this to construct scattering solutions in a neighbourhood of timelike infinity ($i_+$), provided the data on null infinity ($\scri$) decay polynomially. Moreover, the solutions we construct are conormal on a blow-up of Minkowski…
▽ More
We study the energy critical wave equation in 3 dimensions around a single soliton. We obtain energy boundedness (modulo unstable modes) for the linearised problem. We use this to construct scattering solutions in a neighbourhood of timelike infinity ($i_+$), provided the data on null infinity ($\scri$) decay polynomially. Moreover, the solutions we construct are conormal on a blow-up of Minkowski space. The methods of proof also extend to some energy supercritical modifications of the equation.
△ Less
Submitted 20 March, 2024;
originally announced March 2024.
-
Small data nonlinear wave equation numerology: The role of asymptotics
Authors:
Istvan Kadar
Abstract:
Systems of wave equations may fail to be globally well posed, even for small initial data. Attempts to classify systems into well and ill-posed categories work by identifying structural properties of the equations that can work as indicators of well-posedness. The most famous of these are the null and weak null conditions. As noted by Keir, related formulations may fail to properly capture the eff…
▽ More
Systems of wave equations may fail to be globally well posed, even for small initial data. Attempts to classify systems into well and ill-posed categories work by identifying structural properties of the equations that can work as indicators of well-posedness. The most famous of these are the null and weak null conditions. As noted by Keir, related formulations may fail to properly capture the effect of undifferentiated terms in systems of wave equations. We show that this is because null conditions are good for categorising behaviour close to null infinity, but not at timelike infinity. In this paper, we propose an alternative condition for semilinear equations that work for undifferentiated non-linearities as well. We illustrate the strength of this new condition by proving global well and ill-posedness statements for some systems of equation that are not critical according to the our classification. Furthermore, we given two examples of systems satisfying the weak null condition with global ill-posedness due to undifferentiated terms, thereby disproving the weak null conjecture as stated in [DP18].
△ Less
Submitted 14 February, 2023;
originally announced February 2023.
-
On global behaviour of classical effective field theories
Authors:
Istvan Kadar
Abstract:
We continue the rigorous study of classical effective field theories (EFTs) that was recently initiated in the work of Reall and Warnick [RW22]. We study a system with one light and one heavy field with cubic coupling and prove global existence (of the UV solution) under an effective norm in the high mass limit. Furthermore, we prove that the global solution linearly scatters (in Lax-Philips sense…
▽ More
We continue the rigorous study of classical effective field theories (EFTs) that was recently initiated in the work of Reall and Warnick [RW22]. We study a system with one light and one heavy field with cubic coupling and prove global existence (of the UV solution) under an effective norm in the high mass limit. Furthermore, we prove that the global solution linearly scatters (in Lax-Philips sense) and that this final state has an expansion in inverse powers of the mass, and can be recovered from the EFT equation alone.
△ Less
Submitted 19 August, 2022;
originally announced August 2022.
-
Accurate Visual Localization for Automotive Applications
Authors:
Eli Brosh,
Matan Friedmann,
Ilan Kadar,
Lev Yitzhak Lavy,
Elad Levi,
Shmuel Rippa,
Yair Lempert,
Bruno Fernandez-Ruiz,
Roei Herzig,
Trevor Darrell
Abstract:
Accurate vehicle localization is a crucial step towards building effective Vehicle-to-Vehicle networks and automotive applications. Yet standard grade GPS data, such as that provided by mobile phones, is often noisy and exhibits significant localization errors in many urban areas. Approaches for accurate localization from imagery often rely on structure-based techniques, and thus are limited in sc…
▽ More
Accurate vehicle localization is a crucial step towards building effective Vehicle-to-Vehicle networks and automotive applications. Yet standard grade GPS data, such as that provided by mobile phones, is often noisy and exhibits significant localization errors in many urban areas. Approaches for accurate localization from imagery often rely on structure-based techniques, and thus are limited in scale and are expensive to compute. In this paper, we present a scalable visual localization approach geared for real-time performance. We propose a hybrid coarse-to-fine approach that leverages visual and GPS location cues. Our solution uses a self-supervised approach to learn a compact road image representation. This representation enables efficient visual retrieval and provides coarse localization cues, which are fused with vehicle ego-motion to obtain high accuracy location estimates. As a benchmark to evaluate the performance of our visual localization approach, we introduce a new large-scale driving dataset based on video and GPS data obtained from a large-scale network of connected dash-cams. Our experiments confirm that our approach is highly effective in challenging urban environments, reducing localization error by an order of magnitude.
△ Less
Submitted 1 May, 2019;
originally announced May 2019.
