Evaluation of 3D CNN Semantic Mapping for Rover Navigation

TL;DR

This paper presents a method for creating accurate 3D semantic maps of Martian terrain using stereo images and deep learning, aiding autonomous rover navigation and environment understanding.

Contribution

It introduces a novel pipeline combining semantic segmentation with stereo depth mapping to generate 3D semantic maps for planetary exploration.

Findings

Effective semantic mapping of Martian terrain

Utilizes DeepLabv3+ for image labeling

Evaluated on ESA Katwijk Beach Rover Dataset

Abstract

Terrain assessment is a key aspect for autonomous exploration rovers, surrounding environment recognition is required for multiple purposes, such as optimal trajectory planning and autonomous target identification. In this work we present a technique to generate accurate three-dimensional semantic maps for Martian environment. The algorithm uses as input a stereo image acquired by a camera mounted on a rover. Firstly, images are labeled with DeepLabv3+, which is an encoder-decoder Convolutional Neural Networl (CNN). Then, the labels obtained by the semantic segmentation are combined to stereo depth-maps in a Voxel representation. We evaluate our approach on the ESA Katwijk Beach Planetary Rover Dataset.