METAVerse: Meta-Learning Traversability Cost Map for Off-Road Navigation

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Seo, Junwon, Kim, Taekyung, Ahn, Seongyong, Kwak, Kiho
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866929264319791104
author Seo, Junwon
Kim, Taekyung
Ahn, Seongyong
Kwak, Kiho
author_facet Seo, Junwon
Kim, Taekyung
Ahn, Seongyong
Kwak, Kiho
contents Autonomous navigation in off-road conditions requires an accurate estimation of terrain traversability. However, traversability estimation in unstructured environments is subject to high uncertainty due to the variability of numerous factors that influence vehicle-terrain interaction. Consequently, it is challenging to obtain a generalizable model that can accurately predict traversability in a variety of environments. This paper presents METAVerse, a meta-learning framework for learning a global model that accurately and reliably predicts terrain traversability across diverse environments. We train the traversability prediction network to generate a dense and continuous-valued cost map from a sparse LiDAR point cloud, leveraging vehicle-terrain interaction feedback in a self-supervised manner. Meta-learning is utilized to train a global model with driving data collected from multiple environments, effectively minimizing estimation uncertainty. During deployment, online adaptation is performed to rapidly adapt the network to the local environment by exploiting recent interaction experiences. To conduct a comprehensive evaluation, we collect driving data from various terrains and demonstrate that our method can obtain a global model that minimizes uncertainty. Moreover, by integrating our model with a model predictive controller, we demonstrate that the reduced uncertainty results in safe and stable navigation in unstructured and unknown terrains.
format Preprint
id arxiv_https___arxiv_org_abs_2307_13991
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle METAVerse: Meta-Learning Traversability Cost Map for Off-Road Navigation
Seo, Junwon
Kim, Taekyung
Ahn, Seongyong
Kwak, Kiho
Robotics
Computer Vision and Pattern Recognition
Machine Learning
Autonomous navigation in off-road conditions requires an accurate estimation of terrain traversability. However, traversability estimation in unstructured environments is subject to high uncertainty due to the variability of numerous factors that influence vehicle-terrain interaction. Consequently, it is challenging to obtain a generalizable model that can accurately predict traversability in a variety of environments. This paper presents METAVerse, a meta-learning framework for learning a global model that accurately and reliably predicts terrain traversability across diverse environments. We train the traversability prediction network to generate a dense and continuous-valued cost map from a sparse LiDAR point cloud, leveraging vehicle-terrain interaction feedback in a self-supervised manner. Meta-learning is utilized to train a global model with driving data collected from multiple environments, effectively minimizing estimation uncertainty. During deployment, online adaptation is performed to rapidly adapt the network to the local environment by exploiting recent interaction experiences. To conduct a comprehensive evaluation, we collect driving data from various terrains and demonstrate that our method can obtain a global model that minimizes uncertainty. Moreover, by integrating our model with a model predictive controller, we demonstrate that the reduced uncertainty results in safe and stable navigation in unstructured and unknown terrains.
title METAVerse: Meta-Learning Traversability Cost Map for Off-Road Navigation
topic Robotics
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2307.13991