# Latent dynamics learning on PCC-NS-2 soft robot with two constant curvature segments (test)

Source: [Latent dynamics learning on PCC-NS-2 soft robot with two constant curvature segments (test)](<https://www.sota2.com/research/sota/latent-dynamics-learning-on-pcc-ns-2-soft-robot-with-two-constant-curvature-segments-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:59:58.612Z

Evaluation of model performance on learning the latent dynamics using the test split of the PCC-NS-2 soft robot configuration, which features two constant curvature segments. Metrics include RMSE, PSNR, and SSIM.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Latent dynamics learning |
| Dataset (including any stated split) | PCC-NS-2 soft robot with two constant curvature segments (test) |
| Metric | RMSE |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 8 of 8 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [RMSE](<https://www.sota2.com/research/sota/latent-dynamics-learning-on-pcc-ns-2-soft-robot-with-two-constant-curvature-segments-test.md?metric=RMSE>): Lower is better; unit: SCALAR.

- [PSNR](<https://www.sota2.com/research/sota/latent-dynamics-learning-on-pcc-ns-2-soft-robot-with-two-constant-curvature-segments-test.md?metric=PSNR>): Higher is better; unit: SCALAR.

- [SSIM](<https://www.sota2.com/research/sota/latent-dynamics-learning-on-pcc-ns-2-soft-robot-with-two-constant-curvature-segments-test.md?metric=SSIM>): Higher is better; unit: SCALAR.

## Recorded results

| Method | RMSE | Result date | Report paper |
| --- | --- | --- | --- |
| GRU | 0.0951 | Not recorded | 68fa75391e209131cea668c6 |
| CON-S | 0.0996 | Not recorded | 68fa75391e209131cea668c6 |
| CON-M | 0.1008 | Not recorded | 68fa75391e209131cea668c6 |
| MECH-NODE | 0.1035 | Not recorded | 68fa75391e209131cea668c6 |
| CFA-CON | 0.1124 | Not recorded | 68fa75391e209131cea668c6 |
| RNN | 0.1373 | Not recorded | 68fa75391e209131cea668c6 |
| NODE | 0.1867 | Not recorded | 68fa75391e209131cea668c6 |
| coRNN | 0.2504 | Not recorded | 68fa75391e209131cea668c6 |

### GRU

RMSE: 0.0951. Result ID: 699395293a6c314849170557.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75391e209131cea668c6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa617b1e209131ce9f462d.md>)

Recorded evaluation configuration: {"# Parameters":928,"Train. steps per second":1.83,"Inf. time \[ms\]":3.2,"Latent dimension (nz)":8,"Batch size":1,"GPU":"Nvidia RTX 3090"}. This record does not assert independent reproduction.

### CON-S

RMSE: 0.0996. Result ID: 699395293a6c31484917055b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75391e209131cea668c6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75391e209131cea668c6.md>)

Recorded evaluation configuration: {"Model scale":"small","# Parameters":676,"Train. steps per second":0.78,"Inf. time \[ms\]":50.2,"Latent dimension (nz)":8,"Batch size":1,"GPU":"Nvidia RTX 3090"}. This record does not assert independent reproduction.

### CON-M

RMSE: 0.1008. Result ID: 699395293a6c31484917055c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75391e209131cea668c6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75391e209131cea668c6.md>)

Recorded evaluation configuration: {"Model scale":"medium","# Parameters":7048,"Train. steps per second":0.6,"Inf. time \[ms\]":60.1,"Latent dimension (nz)":8,"Batch size":1,"GPU":"Nvidia RTX 3090"}. This record does not assert independent reproduction.

### MECH-NODE

RMSE: 0.1035. Result ID: 699395293a6c31484917055a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75391e209131cea668c6.md>)

Recorded evaluation configuration: {"# Parameters":3062,"Train. steps per second":0.79,"Inf. time \[ms\]":50.3,"Latent dimension (nz)":8,"Batch size":1,"GPU":"Nvidia RTX 3090"}. This record does not assert independent reproduction.

### CFA-CON

RMSE: 0.1124. Result ID: 699395293a6c31484917055d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75391e209131cea668c6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75391e209131cea668c6.md>)

Recorded evaluation configuration: {"# Parameters":7048,"Train. steps per second":1.12,"Inf. time \[ms\]":13.6,"Latent dimension (nz)":8,"Batch size":1,"GPU":"Nvidia RTX 3090"}. This record does not assert independent reproduction.

### RNN

RMSE: 0.1373. Result ID: 699395293a6c314849170556.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75391e209131cea668c6.md>)

Recorded evaluation configuration: {"# Parameters":320,"Train. steps per second":1.87,"Inf. time \[ms\]":2.6,"Latent dimension (nz)":8,"Batch size":1,"GPU":"Nvidia RTX 3090"}. This record does not assert independent reproduction.

### NODE

RMSE: 0.1867. Result ID: 699395293a6c314849170559.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75391e209131cea668c6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66cb1e209131cea12502.md>)

[Implementation](<https://www.sota2.com/github.com/rtqichen/torchdiffeq>)

Recorded evaluation configuration: {"# Parameters":3856,"Train. steps per second":0.79,"Inf. time \[ms\]":50.2,"Latent dimension (nz)":8,"Batch size":1,"GPU":"Nvidia RTX 3090"}. This record does not assert independent reproduction.

### coRNN

RMSE: 0.2504. Result ID: 699395293a6c314849170558.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75391e209131cea668c6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa697a1e209131cea22495.md>)

Recorded evaluation configuration: {"# Parameters":152,"Train. steps per second":1.89,"Inf. time \[ms\]":2.7,"Latent dimension (nz)":8,"Batch size":1,"GPU":"Nvidia RTX 3090"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
