Image-based scene recognition for robot navigation considering traversable plants and its manual annotation-free training

TL;DR

This paper introduces a novel image-based scene recognition framework for robot navigation in plant-rich environments, utilizing unsupervised domain adaptation and traversal experience data to avoid manual annotation.

Contribution

The proposed model combines semantic segmentation and traversability estimation, trained without manual labels, enabling accurate navigation through flexible, plant-covered paths.

Findings

Outperforms conventional semantic segmentation in recognizing traversable plants

Successfully navigates robots in real-world plant-rich environments

Achieves higher accuracy in pixel-wise traversability estimation

Abstract

This paper describes a method of estimating the traversability of plant parts covering a path and navigating through them for mobile robots operating in plant-rich environments. Conventional mobile robots rely on scene recognition methods that consider only the geometric information of the environment. Those methods, therefore, cannot recognize paths as traversable when they are covered by flexible plants. In this paper, we present a novel framework of image-based scene recognition to realize navigation in such plant-rich environments. Our recognition model exploits a semantic segmentation branch for general object classification and a traversability estimation branch for estimating pixel-wise traversability. The semantic segmentation branch is trained using an unsupervised domain adaptation method and the traversability estimation branch is trained with label images generated from the robot's traversal experience during the data acquisition phase, coined traversability masks. The training procedure of the entire model is, therefore, free from manual annotation. In our experiment, we show that the proposed recognition framework is capable of distinguishing traversable plants more accurately than a conventional semantic segmentation with traversable plant and non-traversable plant classes, and an existing image-based traversability estimation method. We also conducted a real-world experiment and confirmed that the robot with the proposed recognition method successfully navigated in plant-rich environments.