End-to-end Differentiable Model of Robot-terrain Interactions

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F24%3A00376781" target="_blank" >RIV/68407700:21230/24:00376781 - isvavai.cz</a>

Výsledek na webu

<a href="https://differentiable.xyz/papers-2024/paper_30.pdf" target="_blank" >https://differentiable.xyz/papers-2024/paper_30.pdf</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

End-to-end Differentiable Model of Robot-terrain Interactions

Popis výsledku v původním jazyce

We propose a differentiable model of robot-terrain interactions that delivers the expected robot trajectory given an onboard camera image and the robot control. The model is trained on a real dataset that covers various terrains ranging from vegetation to man-made obstacles. Since robot-endangering interactions are naturally absent in real-world training data, the consequent learning of the model suffers from training/testing distribution mismatch, and the quality of the result strongly depends on generalization of the model. Consequently, we propose a grey-box, explainable, physics-aware, and end-to-end differentiable model that achieves better generalization through strong geometrical and physical priors. Our model, which functions as an image-conditioned differentiable simulation, can generate millions of trajectories per second and provides interpretable intermediate outputs that enable efficient self-supervision. Our experimental evaluation demonstrates that the model outperforms state-of-the-art methods.

Název v anglickém jazyce

End-to-end Differentiable Model of Robot-terrain Interactions

Popis výsledku anglicky

We propose a differentiable model of robot-terrain interactions that delivers the expected robot trajectory given an onboard camera image and the robot control. The model is trained on a real dataset that covers various terrains ranging from vegetation to man-made obstacles. Since robot-endangering interactions are naturally absent in real-world training data, the consequent learning of the model suffers from training/testing distribution mismatch, and the quality of the result strongly depends on generalization of the model. Consequently, we propose a grey-box, explainable, physics-aware, and end-to-end differentiable model that achieves better generalization through strong geometrical and physical priors. Our model, which functions as an image-conditioned differentiable simulation, can generate millions of trajectories per second and provides interpretable intermediate outputs that enable efficient self-supervision. Our experimental evaluation demonstrates that the model outperforms state-of-the-art methods.

Druh

O - Ostatní výsledky

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

Výsledek vznikl pri realizaci vícero projektů. Více informací v záložce Projekty.

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2024

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů