Experimental investigation of pose informed reinforcement learning for skid-steered visual navigation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Salvi, Ameya, Krovi, Venkat
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911265365950464
author Salvi, Ameya
Krovi, Venkat
author_facet Salvi, Ameya
Krovi, Venkat
contents Vision-based lane keeping is a topic of significant interest in the robotics and autonomous ground vehicles communities in various on-road and off-road applications. The skid-steered vehicle architecture has served as a useful vehicle platform for human controlled operations. However, systematic modeling, especially of the skid-slip wheel terrain interactions (primarily in off-road settings) has created bottlenecks for automation deployment. End-to-end learning based methods such as imitation learning and deep reinforcement learning, have gained prominence as a viable deployment option to counter the lack of accurate analytical models. However, the systematic formulation and subsequent verification/validation in dynamic operation regimes (particularly for skid-steered vehicles) remains a work in progress. To this end, a novel approach for structured formulation for learning visual navigation is proposed and investigated in this work. Extensive software simulations, hardware evaluations and ablation studies now highlight the significantly improved performance of the proposed approach against contemporary literature.
format Preprint
id arxiv_https___arxiv_org_abs_2506_21732
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Experimental investigation of pose informed reinforcement learning for skid-steered visual navigation
Salvi, Ameya
Krovi, Venkat
Robotics
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
Systems and Control
Vision-based lane keeping is a topic of significant interest in the robotics and autonomous ground vehicles communities in various on-road and off-road applications. The skid-steered vehicle architecture has served as a useful vehicle platform for human controlled operations. However, systematic modeling, especially of the skid-slip wheel terrain interactions (primarily in off-road settings) has created bottlenecks for automation deployment. End-to-end learning based methods such as imitation learning and deep reinforcement learning, have gained prominence as a viable deployment option to counter the lack of accurate analytical models. However, the systematic formulation and subsequent verification/validation in dynamic operation regimes (particularly for skid-steered vehicles) remains a work in progress. To this end, a novel approach for structured formulation for learning visual navigation is proposed and investigated in this work. Extensive software simulations, hardware evaluations and ablation studies now highlight the significantly improved performance of the proposed approach against contemporary literature.
title Experimental investigation of pose informed reinforcement learning for skid-steered visual navigation
topic Robotics
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
Systems and Control
url https://arxiv.org/abs/2506.21732