# leggedrobotics/elevation_mapping_cupy

Elevation Mapping on GPU.

Repository: https://github.com/leggedrobotics/elevation_mapping_cupy
Canonical: https://ross.abutalabs.com/products/elevation_mapping_cupy
Language: Python
License: MIT
License Family: permissive
Last push: 2026-08-25T15:47:42+00:00

## Health v2 (maintenance only)
Score: 89/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 69, longevity 100
- inputs: {"age_days": 1623, "days_push": 8, "days_rel": 44, "gap_med": 167.0, "n_releases_24m": 3}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1059, forks 196 (observed 2026-08-28T04:03:25.531275+00:00)

## What it is
A GPU-accelerated elevation mapping library for robotics, built on CuPy and integrated with ROS, that fuses point clouds into multi-modal terrain maps. It provides height drift compensation, raycasting-based artifact removal, and learning-based traversability estimation for legged robot locomotion and navigation.

## Use cases
- build elevation maps from lidar or depth point clouds on GPU
- estimate terrain traversability for legged robot path planning
- compensate state estimation drift in robot terrain mapping
- remove mapping artifacts from overhanging obstacles using raycasting
- fuse geometry, semantics, and RGB data into multi-modal elevation maps
- run real-time terrain perception for quadruped locomotion

## When to choose
- you need real-time elevation mapping for legged robots or mobile robots
- your ROS pipeline needs GPU-accelerated point cloud registration and raycasting
- you want multi-modal terrain layers like traversability, semantics, or RGB
- you are doing research on rough-terrain locomotion and navigation

## When to avoid
- you need mapping on CPU-only hardware without an NVIDIA GPU
- your project does not use ROS or robotics middleware
- you need general-purpose SLAM rather than height-map terrain representation
- you want a plug-and-play mapping solution without robotics integration work

## Facets
- artifact type: library
- maturity: active
- function: computer-vision, image-processing, machine-learning, simulation, plugin-system
- domain: robotics, autonomous-vehicles, computer-vision, machine-learning
- platform: python
- tags: elevation-mapping, cupy, ros, legged-robots, terrain-perception, point-cloud, traversability, gpu-accelerated, linux, gpu

## Member repositories
- leggedrobotics/elevation_mapping_cupy (main) score 89

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:25.531275+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:57:15.249365+00:00, confidence not recorded.
  - readme: https://github.com/leggedrobotics/elevation_mapping_cupy (fetched 2026-08-28T04:03:25.531275+00:00, sha 8caab1ac2423)
- Data as of 2026-08-30T08:39:29.467469+00:00.