elevation_mapping_cupy

Software
Navigation and mapping
Library

A ROS package that builds GPU-accelerated elevation maps from point clouds, aimed at legged robot navigation and locomotion. It can also fuse semantic and RGB data into the map.

Stale

Made by ETH Zurich Robotic Systems Lab, Switzerland (checked on github.com/leggedrobotics (opens in a new tab)).

Source on GitHub: github.com/leggedrobotics/elevation_mapping_cupy (opens in a new tab). Facts read at commit 20a8a26 (opens in a new tab).

Licenses

Code: MIT (opens in a new tab)
Read by Robotheca from LICENSE (opens in a new tab) (MIT) at commit 20a8a26 (opens in a new tab), on 16 Oct 2026.
Commercial use: Yes

Labels

StaleNo commit in 12 months: the newest is from 14 May 2025.

What it does

Compensates for height drift caused by state estimation errors to reduce mapping artifacts.
Uses raycasting and an exclusion zone to remove virtual artifacts and handle overhanging obstacles.
Includes a learning-based traversability filter that scores terrain from local geometry.
Provides smoothing filters and plane segmentation for locomotion over varied terrain.
Supports a multi-modal elevation map that fuses geometry, semantics and RGB from multi-modal point clouds and images.
Runs point cloud registration and ray casting on the GPU for real-time processing.

Written by Robotheca from the project's README (opens in a new tab) at commit 20a8a26.

Getting started

Difficulty: Advanced
How to start: The project's README (opens in a new tab)

Maintenance

Status: Stale: no commit in 12 months
Newest commit: 14 May 2025 on main
Latest release: v2.2.0 - ROS 2 Jazzy GPU correctness and throughput (20 Jul 2026)

Checked on GitHub on 16 Oct 2026.

Robots on Robotheca that use it

No robot on Robotheca is linked to it yet. Links come from a robot's robot.yaml, from its dependency files read as text, or from Robotheca's editors.

Robotheca lists software as a reference: nothing is hosted or run here, and every fact links to where it was read.