Skip to content

Planning-Aligned Pretraining of BEV Representations with Sparse Action-Conditioned Targets for End-to-End Autonomous Driving

Sep 2026 · arXiv.org · Vol abs/2609.22868 · 0 citations
Computer Science

TL;DR

PAVER, Planning-Aligned BEV Encoder Pretraining is introduced, where from a single LiDAR sweep, PAVER constructs sparse risk and unknown targets describing occupied and unobserved evidence along rule-based ego motions, preserving the downstream architecture and camera-only inference.

Abstract

End-to-end driving requires planning-relevant bird's-eye-view (BEV) representations, but existing pretraining approaches often rely on task annotations or dense scene reconstruction. We introduce PAVER, Planning-Aligned BEV Encoder Pretraining. From a single LiDAR sweep, PAVER constructs sparse risk and unknown targets describing occupied and unobserved evidence along rule-based ego motions. A 10K-parameter head predicts these targets from masked BEV features conditioned on the action state, directing supervision toward geometric constraints on candidate motions. Pretraining requires no driving-task annotations or dense reconstruction. Only the BEV encoder is transferred, preserving the downstream architecture and camera-only inference. On nuScenes, PAVER reduces VAD-Tiny's average collision rate from 0.51% to 0.19%, while improving planning L2, motion prediction, detection, and mapping. The selected VAD-Tiny and VAD-Base schedules use about 36% less estimated total training time than scratch training, including pretraining. On Bench2Drive Town05 Long, PAVER improves UniAD-Tiny's closed-loop Driving Score from 48.45 to 58.79. The project page is available at https://archiiive99.github.io/PAVER.

View source

Similar papers

#artificial intelligence Preprint Sep 2026

S2Planner: Multi-Scale Semantic Planner for End-to-End Autonomous Driving

S2Planner is presented, a trajectory planner that combines three front-facing cameras with ego-motion history and the current driving command, and a coarse-to-fine decoder then uses trajectory self-attention and camera-projected cross-attention to refine candidate waypoints.

Zhao-Wei Lu, Li-Guo Zhou, Yu-Jie Guo et al. · 0 citations
Preprint Sep 2026

SV-WAM: An Efficient Surround-View World-Action Model for End-to-End Autonomous Driving

SV-WAM is proposed, a surround-view world-action model (WAM) that preserves full six-camera observations while maintaining efficient inference and introduces a differentiable drivable-area compliance regularizer that penalizes vehicle-footprint corners approaching or crossing drivable boundaries, improving planning saf...

Jin-Yang Wang, Shi-Wei Li, Jun-Jian Wang et al. · 1 citation
Sep 2026

PASSAGE: Scaling Scene-Aligned Motion Learning for Perceptive Humanoid Traversal in Cluttered Environments

PASSAGE, a perception-conditioned planner--tracker framework for humanoid traversal, using virtual reality and inertial motion capture to collect 100 h of scene-aligned human motion across 1,500 cluttered scenes and tests across 50 unseen physical layouts demonstrate traversal without prebuilt maps or offboard computat...

Yuxuan Ma, Zi-Cheng Zeng, Chun-Li Peng et al. · 1 citation
Preprint Sep 2026

READ: Learning Risk-Informed Fields for End-to-End Autonomous Driving

Autonomous driving requires more than recognizing what is present in a scene: a planner must determine how road structure, surrounding agents, and their motion states should influence a future maneuver. Existing learning-based planners can capture these influences through latent scene features and trajectory decoders,...

Zhi-Yuan Liu, Yuan-Xin Tian, Ze-Hong Ke et al. · 0 citations
Preprint Aug 2026

SimWAM: A Simple World Action Model for End-to-End Autonomous Driving

SimWAM, a simple yet effective WAM that leverages future-video prediction as a training-time supervision signal, co-trains a pretrained video expert and a lightweight action expert with joint flow matching and applies reinforcement learning to optimize a compositional driving reward beyond trajectory imitation.

Zongchuang Zhao, Xin Zhou, Tianyang Xu et al. · 4 citations
Sep 2026

ReDrive: Shaping Representations with World Modeling for End-to-End Driving

Driving policies require capabilities of scene understanding and future evolution prediction. To achieve this goal, current end-to-end models typically construct complex perception-planning pipelines or introduce world models that explicitly predict future states, resulting in a complex system architecture. Inspired by...

Yue-Ting Zhu, Shao-Yu Chen, Yue-Hao Song et al. · 0 citations

Related blog posts

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.