Downloads · 30 days
0
ram-shreyas-naik-sabavat/SidewalkPilot-v4.0f
SidewalkPilot-v4.0f is a robotics model from ram-shreyas-naik-sabavat. Use it for the robotics task on the model card, and read the license before you ship it in a product. It is set up for pytorch. The card lists the license as apache-2.0.
SidewalkPilot-v4.0f is a Series 4.0 CF steering model for the SidewalkPilot RC car. Its training contract uses image-only input with current and three future targets used as training supervision. It is the final-epoch…
Downloads · 30 days
0
Access
Public
Updated Jul 19, 2026
Repo size
22.2 MB
Likes
0
Public
Click a slice to open those files.
.onnx22.2 MB · 100%
From the Hugging Face model README
SidewalkPilot-v4.0f is a Series 4.0 CF steering model for the SidewalkPilot RC car. Its training contract uses image-only input with current and three future targets used as training supervision. It is the final-epoch checkpoint paired with v4.0g.
SidewalkPilot-v4.0f.onnx (FP32 ONNX)image [batch,3,180,320]320x180, then (x / 255 - 0.5) / 0.5[batch,4,18]; each horizon contains 9 class logits and 9 class-local offsetsclass_id = argmax(softmax(horizon[0:9]))
fraction = sigmoid(horizon[9 + class_id])
steering_deg = bucket_low[class_id] + fraction * bucket_width[class_id]
Logical steering uses 0 for full left, 90 for straight, and 180 for full right. Future targets are training labels only. They are never runtime inputs.
| Model | Bal9 | Turn exact | Turn +/-1 | ST exact | MAE | Median AE | Signed error |
|---|---|---|---|---|---|---|---|
4.0p | 34.5% | 32.1% | 65.9% | 67.7% | 12.396 | 2.967 | +0.120 |
4.0r | 32.9% | 27.4% | 62.6% | 77.6% | 11.636 | 1.846 | -1.136 |
4.0f (this model) | 25.4% | 23.5% | 56.4% | 62.8% | 15.623 | 6.723 | +1.057 |
4.0g | 20.4% | 17.1% | 46.4% | 76.0% | 14.116 | 2.114 | -1.864 |
4.0a | 33.5% | 30.9% | 65.3% | 68.1% | 12.379 | 3.115 | +0.290 |
4.0c | 32.0% | 29.4% | 62.9% | 75.5% | 11.321 | 1.825 | -0.981 |
v4.0f was the only viable Series 4.0 candidate. In the supervised comparison, v4.0f passed two cases that v3.4 failed, while v3.4 passed two cases that v4.0f failed. The result did not establish either model as generally better, so v3.4 remained the default.
The comparison was supervised and video was captured, but it was not a formal route-controlled benchmark with a preserved per-case score sheet. The claim is limited to the cases presented.
This model does not identify obstacles, prove a clear path, estimate confidence, or detect out-of-distribution scenes. The Series 4.0 history models can amplify their own earlier predictions during closed-loop driving. Do not treat an offline score or successful ONNX load as approval for autonomous use. Keep independent LiDAR emergency braking, manual takeover, conservative speed, and bounded supervised testing above model output.
artifact_manifest.json records the model SHA-256, tensor signature, checkpoint role, field status, and common-set metrics.