RAX: A Robot Cognitive Architecture That Turns Exploration into Persistent Skills
du_yilun · x · 2026-10-17
RAX proposes a cognitive architecture that converts online exploration into persistent robot skills.
- Perception & decisions: The agent uses RGB-D and force observations to build task-relevant 3D models and searches for actions through the robot's control and kinematics tools, developing its own perception pipelines, manipulation strategies, and execution checks.
- Skill consolidation: As exploration scales, successful decisions become autonomous programs containing the full perception–planning–control loop; they run directly, and hand control back to the agent when facing unfamiliar situations.
- Memory: Notes preserve discoveries within a session, while a versioned skill repository stores working code, preconditions, and execution evidence across sessions.
- Parallelism: Exploration is parallelized across agents and environments, cutting time to target performance.
The author clarifies that what scales is the agent's budget of online exploration episodes—execute, observe, revise perception/control code, repeat—while the underlying model stays frozen; learning happens in the agent's growing workspace of code, notes, and verified skills.
More from Embodied
- Google Gemma hardware hackathon at AGI House offers $45k prizes, $15k for LiteRT track — agihouse_org · 2026-10-17
- iPhone Duo pre-orders sell out instantly, shipping slips past January 2027 — mitchdeg · 2026-10-17
- P-BFM: unsupervised RL lets Unitree G1 humanoids adapt to rough terrain from flat-ground references — ChongZzZhang · 2026-10-17
- Researcher builds generative machine vision pipeline reconstructing scenes from monocular video — RexDouglass · 2026-10-17
- Starkey Launches Omega AI+ Hearing Aid With Quad DNN Architecture and Sleep Tracking — BrandonSawalich · 2026-10-17
- RAX open-sources everything: zero demos, all experiment sessions included — du_yilun · 2026-10-17