# Robot Manipulation on RLBench-OG (Robustness Evaluation)

Source: [Robot Manipulation on RLBench-OG (Robustness Evaluation)](<https://www.sota2.com/research/sota/robot-manipulation-on-rlbench-og-robustness-evaluation>)

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

Data updated: 2026-03-19T08:47:21.804Z

Evaluation of robot manipulation success rates and sensitivity across various environmental conditions (occlusion, lighting, texture, distractors, and camera pose) using the RLBench-OG dataset.

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 | Robot Manipulation |
| Dataset (including any stated split) | RLBench-OG |
| Metric | Average Success Rate |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [Average Success Rate](<https://www.sota2.com/research/sota/robot-manipulation-on-rlbench-og-robustness-evaluation.md?metric=Average%20Success%20Rate>): Higher is better; unit: PERCENT.

- [Average Rank](<https://www.sota2.com/research/sota/robot-manipulation-on-rlbench-og-robustness-evaluation.md?metric=Average%20Rank>): Lower is better; unit: SCALAR.

- [Success Rate (Occlusion 1)](<https://www.sota2.com/research/sota/robot-manipulation-on-rlbench-og-robustness-evaluation.md?metric=Success%20Rate%20(Occlusion%201)>): Higher is better; unit: PERCENT.

- [Success Rate (Occlusion 2)](<https://www.sota2.com/research/sota/robot-manipulation-on-rlbench-og-robustness-evaluation.md?metric=Success%20Rate%20(Occlusion%202)>): Higher is better; unit: PERCENT.

- [Success Rate (Light Color)](<https://www.sota2.com/research/sota/robot-manipulation-on-rlbench-og-robustness-evaluation.md?metric=Success%20Rate%20(Light%20Color)>): Higher is better; unit: PERCENT.

- [Success Rate (Table Color)](<https://www.sota2.com/research/sota/robot-manipulation-on-rlbench-og-robustness-evaluation.md?metric=Success%20Rate%20(Table%20Color)>): Higher is better; unit: PERCENT.

- [Success Rate (Table Texture)](<https://www.sota2.com/research/sota/robot-manipulation-on-rlbench-og-robustness-evaluation.md?metric=Success%20Rate%20(Table%20Texture)>): Higher is better; unit: PERCENT.

- [Success Rate (Distractor)](<https://www.sota2.com/research/sota/robot-manipulation-on-rlbench-og-robustness-evaluation.md?metric=Success%20Rate%20(Distractor)>): Higher is better; unit: PERCENT.

- [Success Rate (Background Texture)](<https://www.sota2.com/research/sota/robot-manipulation-on-rlbench-og-robustness-evaluation.md?metric=Success%20Rate%20(Background%20Texture)>): Higher is better; unit: PERCENT.

- [Success Rate (Camera Pose)](<https://www.sota2.com/research/sota/robot-manipulation-on-rlbench-og-robustness-evaluation.md?metric=Success%20Rate%20(Camera%20Pose)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Average Success Rate | Result date | Report paper |
| --- | --- | --- | --- |
| TVVE | 67 | Not recorded | 68fa7c691e209131cea8dece |
| RVT2 | 64.5 | Not recorded | 68fa7c691e209131cea8dece |
| ARP | 63.7 | Not recorded | 68fa7c691e209131cea8dece |
| Diffusion Policy | 23.8 | Not recorded | 68fa7c691e209131cea8dece |

### TVVE

Average Success Rate: 67. Result ID: 69bbb81647f34ce924a5a4b0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c691e209131cea8dece.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c691e209131cea8dece.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### RVT2

Average Success Rate: 64.5. Result ID: 69bbb81647f34ce924a5a4af.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c691e209131cea8dece.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73931e209131cea5d546.md>)

[Implementation](<https://robotic-view-transformer-2.github.io/>)

Recorded evaluation configuration: {"Venue":"RSS"}. This record does not assert independent reproduction.

### ARP

Average Success Rate: 63.7. Result ID: 69bbb81647f34ce924a5a4ae.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c691e209131cea8dece.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77a01e209131cea7474f.md>)

[Implementation](<http://github.com/mlzxy/arp>)

Recorded evaluation configuration: {"Venue":"RA-L"}. This record does not assert independent reproduction.

### Diffusion Policy

Average Success Rate: 23.8. Result ID: 69bbb81647f34ce924a5a4ad.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c691e209131cea8dece.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72841e209131cea56f68.md>)

Recorded evaluation configuration: {"Venue":"IJRR"}. This record does not assert independent reproduction.

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