Saved in:
Bibliographic Details
Main Authors: Buoso, Davide, Robinson, Luke, Averta, Giuseppe, Torr, Philip, Franzmeyer, Tim, De Martini, Daniele
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2411.04006
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910686900125696
author Buoso, Davide
Robinson, Luke
Averta, Giuseppe
Torr, Philip
Franzmeyer, Tim
De Martini, Daniele
author_facet Buoso, Davide
Robinson, Luke
Averta, Giuseppe
Torr, Philip
Franzmeyer, Tim
De Martini, Daniele
contents This study explores the potential of off-the-shelf Vision-Language Models (VLMs) for high-level robot planning in the context of autonomous navigation. Indeed, while most of existing learning-based approaches for path planning require extensive task-specific training/fine-tuning, we demonstrate how such training can be avoided for most practical cases. To do this, we introduce Select2Plan (S2P), a novel training-free framework for high-level robot planning which completely eliminates the need for fine-tuning or specialised training. By leveraging structured Visual Question-Answering (VQA) and In-Context Learning (ICL), our approach drastically reduces the need for data collection, requiring a fraction of the task-specific data typically used by trained models, or even relying only on online data. Our method facilitates the effective use of a generally trained VLM in a flexible and cost-efficient way, and does not require additional sensing except for a simple monocular camera. We demonstrate its adaptability across various scene types, context sources, and sensing setups. We evaluate our approach in two distinct scenarios: traditional First-Person View (FPV) and infrastructure-driven Third-Person View (TPV) navigation, demonstrating the flexibility and simplicity of our method. Our technique significantly enhances the navigational capabilities of a baseline VLM of approximately 50% in TPV scenario, and is comparable to trained models in the FPV one, with as few as 20 demonstrations.
format Preprint
id arxiv_https___arxiv_org_abs_2411_04006
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Select2Plan: Training-Free ICL-Based Planning through VQA and Memory Retrieval
Buoso, Davide
Robinson, Luke
Averta, Giuseppe
Torr, Philip
Franzmeyer, Tim
De Martini, Daniele
Robotics
Artificial Intelligence
This study explores the potential of off-the-shelf Vision-Language Models (VLMs) for high-level robot planning in the context of autonomous navigation. Indeed, while most of existing learning-based approaches for path planning require extensive task-specific training/fine-tuning, we demonstrate how such training can be avoided for most practical cases. To do this, we introduce Select2Plan (S2P), a novel training-free framework for high-level robot planning which completely eliminates the need for fine-tuning or specialised training. By leveraging structured Visual Question-Answering (VQA) and In-Context Learning (ICL), our approach drastically reduces the need for data collection, requiring a fraction of the task-specific data typically used by trained models, or even relying only on online data. Our method facilitates the effective use of a generally trained VLM in a flexible and cost-efficient way, and does not require additional sensing except for a simple monocular camera. We demonstrate its adaptability across various scene types, context sources, and sensing setups. We evaluate our approach in two distinct scenarios: traditional First-Person View (FPV) and infrastructure-driven Third-Person View (TPV) navigation, demonstrating the flexibility and simplicity of our method. Our technique significantly enhances the navigational capabilities of a baseline VLM of approximately 50% in TPV scenario, and is comparable to trained models in the FPV one, with as few as 20 demonstrations.
title Select2Plan: Training-Free ICL-Based Planning through VQA and Memory Retrieval
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2411.04006