This page primarily introduces ANIMA, a domain-agnostic six-layer / five-factor cognition framework that turns upstream signals (text, voice, BCI, vision) into verifiable, dispatchable, recoverable robot actions. It ships as a standalone Python library at jeffliulab/anima (v0.1.0).
A shared cognition framework: six-layer stack (L0–L5) + six-gate Test-and-Check + five-factor self-assessment (ITA / MQA / SQA / GOA / PEA). Ships as a pip-installable standalone Python library at jeffliulab/anima starting with v0.1.0.
Not a vision model, motion planner, or hardware driver; not locked to a single robot embodiment. The LLM acts only as a parser — it never emits motor commands directly.
Shipped as a standalone repo on 2026-04-21. Two application branches — soma-arm (tabletop chess) and soma-care (simulated-ward ADL) — share the same framework and only diverge at the L4 adapter layer.
These pages continue the story around architecture, validation, and skill boundaries without overloading the landing page.
ANIMA's L0–L5 stack translates upstream signals (text, voice, BCI, vision) into verifiable robot actions layer by layer, each layer carrying a clear input/output contract and a named failure path.
Test-and-Check is one of ANIMA's defining moves: it turns the claim of task completion into a proposition that must be re-verified. Six gates reject uninterpretable specs before execution; three stages (Pre / Runtime / Post) ensure failure never gets silently papered over.
ANIMA touches concrete robots through the skill registry (L3) and the EmbodiedAdapter protocol (L4), without swallowing hardware-specific details into the framework. This boundary determines whether ANIMA can move to the next robot or stays trapped inside a single demo.
The five factors are the structural core of ANIMA's self-assessment. Each evaluates at Pre / Runtime / Post, and the composite yields a success probability with calibrated confidence. GOA composes multiplicatively, SQA learns from history, and PEA makes the system better over time.
The most concrete reference implementation today
A tabletop language-driven robot arm project used to validate ANIMA's cognition loop, visual grounding, skill execution, and failure recovery.
It is not the final product. It is the benchmark and manipulation capability layer on the path to home robotics.
Medical-care intent-to-action simulation
The medical-care product line of the SOMA family. Powered by the ANIMA cognition framework, it turns intent (text, voice, and future BCI signals) into auditable robot actions inside a simulated hospital ward. v0.4 focuses on showcasing the full intent-to-care loop via MuJoCo simulation and offline video.
The care product line of the SOMA family, running in parallel with soma-arm and sharing the ANIMA cognition framework.