Map, Inspect, and Validate Embodied AI Tools
A public research-engineering hub for robotics and embodied-intelligence tools. Browse perception, mapping, planning, and control modules, inspect their input/output contracts, follow official resources, and track which entries have benchmark evidence or runnable validation paths.
Core Modules
Move from public resources to tools, benchmark signals, and task-oriented discovery.
Featured Tools
Representative entries with source images, paper context, demos, benchmark rows, and implementation links.
Dense Object Nets learn pixel-level object descriptors that let a robot identify corresponding points across views and use them as manipulation targets.
OctoMap is a probabilistic 3D occupancy mapping framework based on octrees, commonly used for collision checking, navigation, and planning around observed space.
ActPerMoMa chooses informative base and camera motions so a mobile manipulator can improve object perception before acting.
Nav2 is the ROS 2 navigation stack for planning, controlling, recovering, and executing autonomous robot navigation tasks.
Capability Coverage
Tools are grouped by embodied-intelligence capability so entries stay easy to review and extend.