# Addressing the Waypoint-Action Gap in End-to-End Autonomous Driving via Vehicle Motion Models

Source: [Addressing the Waypoint-Action Gap in End-to-End Autonomous Driving via Vehicle Motion Models](<https://www.sota2.com/research/paper/69896a27c97a0b3e3b5dc5a5>)

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

Data updated: 2026-02-11T05:00:13.386Z

Authors: Jorge Daniel Rodr\\'iguez-Vidal, Gabriel Villalonga, Diego Porres, Antonio M. L\\'opez Pe\\~na

Published: 2026-02-05T21:42:13.000Z

## Abstract

End-to-End Autonomous Driving (E2E-AD) systems are typically grouped by the nature of their outputs: (i) waypoint-based models that predict a future trajectory, and (ii) action-based models that directly output throttle, steer and brake. Most recent benchmark protocols and training pipelines are waypoint-based, which makes action-based policies harder to train and compare, slowing their progress. To bridge this waypoint-action gap, we propose a novel, differentiable vehicle-model framework that rolls out predicted action sequences to their corresponding ego-frame waypoint trajectories while supervising in waypoint space. Our approach enables action-based architectures to be trained and evaluated, for the first time, within waypoint-based benchmarks without modifying the underlying evaluation protocol. We extensively evaluate our framework across multiple challenging benchmarks and observe consistent improvements over the baselines. In particular, on NAVSIM \\texttt{navhard} our approach achieves state-of-the-art performance. Our code will be made publicly available upon acceptance.

## Recorded benchmark results

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.

- [End-to-End Autonomous Driving on NAVSIM navhard v2 (S1)](<https://www.sota2.com/research/sota/end-to-end-autonomous-driving-on-navsim-navhard-v2-s1.md>): End-to-end Autonomous Driving; NAVSIM navhard v2 (S1); NC: 96.2.

- [End-to-End Autonomous Driving on NAVSIM navhard v2 (S2)](<https://www.sota2.com/research/sota/end-to-end-autonomous-driving-on-navsim-navhard-v2-s2.md>): End-to-end Autonomous Driving; NAVSIM navhard v2 (S2); NC: 80.6.

- [End-to-End Autonomous Driving on NAVSIM navhard v2](<https://www.sota2.com/research/sota/end-to-end-autonomous-driving-on-navsim-navhard-v2.md>): End-to-end Autonomous Driving; NAVSIM navhard v2; EPDMS: 21.

- [Autonomous Driving on NAVSIM (test)](<https://www.sota2.com/research/sota/autonomous-driving-on-navsim-test.md>): Autonomous Driving; NAVSIM (test); PDMS: 84.5.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2602.06214>)

[PDF](<https://arxiv.org/pdf/2602.06214.pdf>)
