Bench2Drive On Bench2Drive

评估指标

Driving Score

评测结果

各个模型在此基准测试上的表现结果

Paper TitleRepository
HiP-AD86.77HiP-AD: Hierarchical and Multi-Granularity Planning with Deformable Attention for Autonomous Driving in a Single Decoder
SimLingo-Base (CarLLaVa)85.94CarLLaVA: Vision language models for camera-only closed-loop driving
TransFuser++84.21Hidden Biases of End-to-End Driving Models
Hydra-NeXt73.86Hydra-NeXt: Robust Closed-Loop Driving with Open-Loop Training
DiffAD67.92DiffAD: A Unified Diffusion Modeling Approach for Autonomous Driving-
DriveAdapter64.22DriveAdapter: Breaking the Coupling Barrier of Perception and Planning in End-to-End Autonomous Driving
Drivetransformer-Large63.46DriveTransformer: Unified Transformer for Scalable End-to-End Autonomous Driving
ThinkTwice62.44Think Twice before Driving: Towards Scalable Decoders for End-to-End Autonomous Driving
TCP-traj59.90Trajectory-guided Control Prediction for End-to-end Autonomous Driving: A Simple yet Strong Baseline
DiFSD52.02DiFSD: Ego-Centric Fully Sparse Paradigm with Uncertainty Denoising and Iterative Refinement for Efficient End-to-End Self-Driving
TCP-traj w/o distillation49.30Trajectory-guided Control Prediction for End-to-end Autonomous Driving: A Simple yet Strong Baseline
UniAD-Base45.81Planning-oriented Autonomous Driving
GenAD44.81GenAD: Generative End-to-End Autonomous Driving
SparseDrive44.54SparseDrive: End-to-End Autonomous Driving via Sparse Scene Representation
VAD42.35VAD: Vectorized Scene Representation for Efficient Autonomous Driving
UniAD-Tiny40.73Planning-oriented Autonomous Driving
TCP40.70Trajectory-guided Control Prediction for End-to-end Autonomous Driving: A Simple yet Strong Baseline
TCP-ctrl30.47Trajectory-guided Control Prediction for End-to-end Autonomous Driving: A Simple yet Strong Baseline
AD-MLP18.05Bench2Drive: Towards Multi-Ability Benchmarking of Closed-Loop End-To-End Autonomous Driving
0 of 19 row(s) selected.
Bench2Drive On Bench2Drive | SOTA | HyperAI超神经