# Advisory autonomy on Advisory Autonomy Highway ramp (Acceleration guidance)

Source: [Advisory autonomy on Advisory Autonomy Highway ramp (Acceleration guidance)](<https://www.sota2.com/research/sota/advisory-autonomy-on-advisory-autonomy-highway-ramp-acceleration-guidance>)

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

Data updated: 2026-02-26T05:53:48.545Z

Evaluation of advisory autonomy performance specifically on the Highway ramp scenario, guided by acceleration parameters, measured using Normalized Reward.

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 | Advisory autonomy |
| Dataset (including any stated split) | Advisory Autonomy Highway ramp (Acceleration guidance) |
| Metric | Normalized Reward |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| 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

- [Normalized Reward](<https://www.sota2.com/research/sota/advisory-autonomy-on-advisory-autonomy-highway-ramp-acceleration-guidance.md?metric=Normalized%20Reward>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Normalized Reward | Result date | Report paper |
| --- | --- | --- | --- |
| Sequential Oracle | 0.712 | Not recorded | 68fa76b71e209131cea6ea9f |
| Random Selection | 0.6599 | Not recorded | 68fa76b71e209131cea6ea9f |
| Greedy Selection | 0.657 | Not recorded | 68fa76b71e209131cea6ea9f |
| MBTL | 0.6282 | Not recorded | 68fa76b71e209131cea6ea9f |
| Multi-task Training | 0.5374 | Not recorded | 68fa76b71e209131cea6ea9f |
| Independent Training | 0.4276 | Not recorded | 68fa76b71e209131cea6ea9f |

### Sequential Oracle

Normalized Reward: 0.712. Result ID: 69944b163a6c3148491c5174.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76b71e209131cea6ea9f.md>)

Recorded evaluation configuration: {"Number of Trained Models":"N"}. This record does not assert independent reproduction.

### Random Selection

Normalized Reward: 0.6599. Result ID: 69944b163a6c3148491c5171.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76b71e209131cea6ea9f.md>)

Recorded evaluation configuration: {"Number of Trained Models":"K"}. This record does not assert independent reproduction.

### Greedy Selection

Normalized Reward: 0.657. Result ID: 69944b163a6c3148491c5172.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76b71e209131cea6ea9f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74db1e209131cea64342.md>)

Recorded evaluation configuration: {"Number of Trained Models":"K"}. This record does not assert independent reproduction.

### MBTL

Normalized Reward: 0.6282. Result ID: 69944b163a6c3148491c5173.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76b71e209131cea6ea9f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76b71e209131cea6ea9f.md>)

Recorded evaluation configuration: {"Number of Trained Models":"K"}. This record does not assert independent reproduction.

### Multi-task Training

Normalized Reward: 0.5374. Result ID: 69944b163a6c3148491c5170.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76b71e209131cea6ea9f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e651e209131cea3fad8.md>)

Recorded evaluation configuration: {"Number of Trained Models":"1"}. This record does not assert independent reproduction.

### Independent Training

Normalized Reward: 0.4276. Result ID: 69944b163a6c3148491c516f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76b71e209131cea6ea9f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e651e209131cea3fad8.md>)

Recorded evaluation configuration: {"Number of Trained Models":"N"}. This record does not assert independent reproduction.

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