publish roller-pollen-reference: perpetual, policy.onnx

README.md

@@ -0,0 +1,30 @@
+---
+tags:
+- microduck
+- robotics
+- reinforcement-learning
+- onnx
+library_name: microduck
+pipeline_tag: robotics
+---
+
+# roller-pollen-reference
+
+roller-pollen-reference
+
+A **perpetual** policy for the [microduck](https://github.com/pollen-robotics/microduck) (61-D observation, 14 actions, 50 Hz, on rollers). Runs until told otherwise — a gait, loaded into a policy slot.
+
+## Run it on a robot
+
+```bash
+sudo robotctl policy load <slot> hunjison/microduck-roller-pollen-reference
+```
+
+The observation normalizer is baked into `policy.onnx`; feed raw observations.
+`manifest.json` follows schema 2 of the microduck policy manifest (`docs/policy-manifest.md` in the daemon repo).
+
+## Training
+
+- **repo**: `pollen-robotics/microduck_rl`
+- **branch**: `main`
+- **commit**: `6ff40704a`

manifest.json

@@ -0,0 +1,41 @@
+{
+  "schema_version": 2,
+  "model_api": 1,
+  "obs_len": 61,
+  "action_len": 14,
+  "robot": {
+    "model": "microduck",
+    "hw_rev": 1,
+    "servos": "xl330",
+    "control_hz": 50,
+    "accessories": [
+      "rollers"
+    ]
+  },
+  "name": "roller-pollen-reference",
+  "kind": "perpetual",
+  "entry_pose": "standing",
+  "description": "roller-pollen-reference",
+  "command": {
+    "encoding": "constant",
+    "idle": [
+      0.0,
+      0.0,
+      0.0
+    ],
+    "twist": "unused (zeros)",
+    "head": "unused (zeros)",
+    "body": "unused (zeros)"
+  },
+  "duration_s": null,
+  "training": {
+    "repo": "pollen-robotics/microduck_rl",
+    "commit": "6ff40704a",
+    "branch": "main",
+    "dirty": false,
+    "source_file": "policy.onnx"
+  },
+  "arena": {
+    "event": "roller-sprint-2m"
+  }
+}

policy.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01edf63a529a8396d46dac62bb427fda1ce67685611e72ddf790bf46f610c82a
+size 793753