MichaelGrupp/evo

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

4309 stars · 795 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

evo is a Python package with CLI executables and a small library for evaluating, comparing, and plotting trajectory output from odometry and SLAM algorithms. It supports TUM, KITTI, EuRoC, and ROS/ROS2 bagfile trajectory formats with alignment, scale adjustment, and metric computation.

Use cases

• evaluate SLAM trajectory accuracy against ground truth
• compare odometry outputs from multiple algorithms
• compute APE and RPE metrics on TUM or KITTI trajectories
• plot and export trajectory error plots for papers
• align and scale monocular SLAM trajectories
• analyze trajectories from ROS or ROS2 bag files

When to choose

• you need to benchmark or evaluate odometry/SLAM trajectories in standard formats
• you want a configurable CLI plus a Python library for custom trajectory analysis
• you work with ROS/ROS2, KITTI, TUM, or EuRoC data

When to avoid

• you need a 1-to-1 reimplementation of a dataset-specific evaluation protocol
• you need real-time SLAM or odometry itself rather than offline evaluation
• you require a non-GPL license for redistribution

Facets

library · maturity stable

benchmarking data-visualization cli math robotics simulation developer-tools data-visualization windows python cli cross-platform slam odometry trajectory-evaluation ros ros2 kitti tum euroc metrics linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="MichaelGrupp/evo")