MeanFuser: Fast One-Step Multi-Modal Trajectory Generation and Adaptive Reconstruction via MeanFlow for End-to-End Autonomous Driving

Junli Wang; Yinan Zheng; Xueyi Liu; Zebin Xing; Pengfei Li; Guang Li; Kun Ma; Guang Chen

cs.CVFeb 23, 2026

MeanFuser: Fast One-Step Multi-Modal Trajectory Generation and Adaptive Reconstruction via MeanFlow for End-to-End Autonomous Driving

Authors: Junli Wang, Yinan Zheng, Xueyi Liu, Zebin Xing, Pengfei Li, Guang Li, Kun Ma, Guang Chen,

Organizations: SKL-MAIS, Institute of Automation, Chinese Academy of Sciences · School of Artificial Intelligence, University of Chinese Academy of Sciences · Xiaomi EV · Institute for AI Industry Research (AIR), Tsinghua University

Abstract

Generative models have shown great potential in trajectory planning. Recent studies demonstrate that anchor-guided generative models are effective in modeling the uncertainty of driving behaviors and improving overall performance. However, these methods rely on discrete anchor vocabularies that must sufficiently cover the trajectory distribution during testing to ensure robustness, inducing an inherent trade-off between vocabulary size and model performance. To overcome this limitation, we propose MeanFuser, an end-to-end autonomous driving method that enhances both efficiency and robustness through three key designs. (1) We introduce Gaussian Mixture Noise (GMN) to guide generative sampling, enabling a continuous representation of the trajectory space and eliminating the dependency on discrete anchor vocabularies. (2) We adapt ``MeanFlow Identity" to end-to-end planning, which models the mean velocity field between GMN and trajectory distribution instead of the instantaneous velocity field used in vanilla flow matching methods, effectively eliminating numerical errors from ODE solvers and significantly accelerating inference. (3) We design a lightweight Adaptive Reconstruction Module (ARM) that enables the model to implicitly select from all sampled proposals or reconstruct a new trajectory when none is satisfactory via attention weights.Experiments on the NAVSIM closed-loop benchmark demonstrate that MeanFuser achieves outstanding performance without the supervision of the PDM Score and exceptional inference efficiency, offering a robust and efficient solution for end-to-end autonomous driving. Our code and model are available at https://github.com/wjl2244/MeanFuser.

Figures & tables

Panel of Figure 1 — Figure 1: (a) illustrates the differences between our proposed method and existing generative approaches, highlighting the introduction of Gaussian mixture noise to replace anchor vocabularies, one-step sampling, and the adaptive reconstruction module. (b) shows the advantages of MeanFuser over GoalFlow [ 33 ] , Hydra-MDP [ 20 ] , and DiffusionDrive [ 22 ] in terms of closed-loop performance, inference speed and plan module inference speed.

Figure 2: Failure scene visualization. Anchor-guided models (GoalFlow, DiffusionDrive) fail due to the inability of discrete vocabularies to cover the entire trajectory space, while our model generates proposals that encompass the optimal trajectories.

Figure 3: Overall architecture of MeanFuser. Training : During training, both the images and ego-vehicle states are encoded into context features, with auxiliary supervision from mapping tasks. The model is conditioned on these context features to learn the average velocity field $u_{\theta}$ over the time interval $r$ and $t$ . Multi-Modal Sample : Noise samples are drawn from Gaussian Mixture Noise, and the one-step sampling formulation is then applied to generate diverse multi-modal trajectories. Adaptive Reconstruction Module : The sampled multi-modal trajectories are encoded and fused with the context features through cross-attention, after which a Projector outputs the final planning trajectory.

Method	Venue	Input	Img. Backbone	NC $\uparrow$	DAC $\uparrow$	TTC $\uparrow$	Comf. $\uparrow$	EP $\uparrow$	PDMS $\uparrow$
TransFuser [ 7 ]	IEEE TPAMI	C & L	ResNet-34	97.7	92.8	92.8	100	79.2	84.0
VADv2 [ 3 ]	arXiv 2024	C & L	ResNet-34	97.2	89.1	91.6	100	76.0	80.9
Hydra-MDP [ 20 ]	arXiv 2024	C & L	ResNet-34	98.3	96.0	94.6	100	78.7	86.5
GoalFlow* [ 33 ]	CVPR 2025	C & L	ResNet-34	98.3	93.8	94.3	100	79.8	85.7
DiffusionDrive [ 22 ]	CVPR 2025	C & L	ResNet-34	98.2	96.2	94.7	100	82.2	88.1
WoTE [ 19 ]	ICCV 2025	C & L	ResNet-34	98.5	96.8	94.9	99.9	81.9	88.3

Table 1: Performance on the NAVSIMv1 navtest benchmark. “C” denotes camera, and “L” denotes LiDAR. * indicates results reported from the official papers.

Method	NC $\uparrow$	DAC $\uparrow$	DDC $\uparrow$	TLC $\uparrow$	EP $\uparrow$	TTC $\uparrow$	LK $\uparrow$	HC $\uparrow$	EC $\uparrow$	EPDMS $\uparrow$
Ego Status MLP [ 21 ]	93.1	77.9	92.7	99.6	86.0	91.5	89.4	98.3	85.4	64.0
TransFuser [ 7 ]	96.9	89.9	97.8	99.7	87.1	95.4	92.7	98.3	87.2	76.7
Hydra-MDP++ [ 18 ]	97.2	97.5	99.4	99.6	83.1	96.5	94.4	98.2	70.9	81.4
DriveSuprim [ 36 ]	97.5	96.5	99.4	99.6	88.4	96.6	95.5	98.3	77.0	83.1
DiffusionDrive † [ 22 ]	98.2	96.3	99.4	99.8	87.4	97.4	97.0	98.3	87.7	88.3
MeanFuser(Ours)	98.3	97.2	99.6	99.8	87.6	97.4	97.3	98.3	88.2	89.5

Table 2: Performance of the NAVSIMv2 navtest benchmark. † denotes testing with the official checkpoint

Method	Dim	Params (M)	PDMS $\uparrow$	Plan FPS $\uparrow$	FPS $\uparrow$
TransFuser	256	55.9	84.0	3934	63
GoalFlow	256	62.3	85.7	11	10
Hydra-MDP	256	65.2	86.5	25	20
DiffusionDrive	256	60.7	88.1	75	39
MeanFuser	128	54.6	89.0	434	59

Table 3: Model parameter size, inference speed, and performance. Bold and underlined values denote the best and second-best results, respectively. “Dim” indicates the number of hidden neurons in the model, “FPS” represents the median inference speed measured on a single NVIDIA H20 GPU over multiple runs, and “Plan FPS” refers to the inference speed of trajectory planning excluding the perception encoder.

Model	Components	PDMS $\uparrow$	$N_{proposals}$	$P_{L_{2}>0.2}$ $\downarrow$	$P_{L_{2}>0.5}$ $\downarrow$	$N_{DAC=0}$ $\downarrow$
-	DiffusionDrive	88.1	20	72.7%	20.0%	84
Base	TransFuser	84.0	-	-	-	-
$\mathcal{M}_{0}$	Base + vanilla MeanFlow	87.3 +3.3	16	92.8%	40.6%	143
$\mathcal{M}_{1}$	$\mathcal{M}_{0}$ + Guassian-mixed Noise ( GMN )	88.2 +0.9	16	69.3%	18.5%	58
$\mathcal{M}_{2}$	$\mathcal{M}_{1}$ + Adpative Reconstruction Module ( ARM )	89.0 +0.8	17	67.1%	16.9%	48
$\mathcal{M}_{3}$	$\mathcal{M}_{1}$ + Trajectory Proposals Averaging	71.2 -17.8	17	69.1%	18.0%	57

Table 4: Ablation study on the impact of each module. Base denotes the TransFuser [ 7 ] baseline.

\mathcal{M}_{0}

replaces the decoder MLP in TransFuser with a MeanFlow model and employs a classifier to score the parallel sampled trajectory proposals.

\mathcal{M}_{1}

and

\mathcal{M}_{2}

further enhance

\mathcal{M}_{0}

by introducing GMN and an ARM, respectively.

\mathcal{M}_{3}

employs an averaging strategy across all sampled traiectories.

N_{\text{proposals}}

denotes the number of sampled trajectory proposals;

P_{L_{2}>0.2}

and

P_{L_{2}>0.5}

represent the proportions of scenarios where all proposals have a minimum

L_{2}

distance greater than 0.2 and 0.5 from the expert trajectory; and

N_{\text{DAC}=0}

indicates the number of cases where all proposals leave the drivable area.

Figure 4: Visualization of sampling from different Gaussian components. Parallel sampling of trajectories from distinct Gaussian components can generate diverse driving styles, ranging from conservative to aggressive.

Figure 5: Visualization of the multimodal trajectory of the model. The left image shows expert demonstration trajectories, while the right image displays the model’s inferred strategies for maintaining a straight path and performing a left lane change, representing two different modes.

Method	RC $\uparrow$	IS $\uparrow$	DS $\uparrow$
Latent TransFuser [ 7 ]	95.18 ±0.45	0.38 ±0.05	37.31 ±5.35
TransFuser [ 7 ]	93.38 ±1.20	0.50 ±0.06	47.30 ±5.72
DiffusionDrive [ 22 ]	94.16 ±1.46	0.69 ±0.02	64.27 ±2.43
MeanFuser (Ours)	94.65 ±1.32	0.73 ±0.05	70.08 ±3.20

Table 5: Longest6 Benchmark Results. We show the mean and std for all metrics (RC: Route Completion, IS: Infraction Score. DS: Driving Score).

Figure 6: Visualization on the CARLA Longest6 benchmark. The two images in the upper row and the two in the lower row showcase the planning outcomes, depicting daytime and nighttime conditions, respectively.

Panel of Figure 7 — Figure 7: Visualization of alternative approaches for generating Gaussian Mixture Noise (GMN) . (a) Mean and standard deviation are derived from clustered expert demonstrations in the training set. (b) Mean and standard deviation are obtained through manually design.

$N_{gaussian}$	NC $\uparrow$	DAC $\uparrow$	TTC $\uparrow$	Comf. $\uparrow$	EP $\uparrow$	PDMS $\uparrow$
2	98.1	96.9	94.0	100	81.8	88.4 -0.5
8	98.6	97.0	95.0	100	82.8	89.0
16	98.1	97.3	93.8	99.9	83.3	88.8 -0.2
32	98.0	97.0	94.3	99.9	82.8	88.5 -0.5

Table 6: Number of Gaussian components and model performance.

N_{gaussian}

denotes the number of Gaussian components in the Gaussian Mixture Noise distribution.

Method	NC $\uparrow$	DAC $\uparrow$	TTC $\uparrow$	Comf. $\uparrow$	EP $\uparrow$	PDMS $\uparrow$
Data Clustering	98.6	97.0	95.0	100	82.8	89.0
Manual Design	98.2	97.0	94.1	100	83.0	88.6 -0.4

Table 7: Comparison of the GMN Generation Methods.

Method	GMN	$K$	PDMS $\uparrow$	$\mathcal{D}$ $\uparrow$	$\mathcal{M}_{DP}$ $\uparrow$
TransFuser	-	8	94.0	0.0	-
MeanFuser	$\times$	8	88.3	0.25	22.07
MeanFuser	$\surd$	8	89.0	0.30	26.70 +20.84%

Table 8: Comparison of multimodality and performance. (GMN: Gaussian Mixture Noise.

K

: number of multimodal trajectories;

\mathcal{D}

: multimodality metric.)

Figure 8: Visualization of the intersection and union dynamics of bounding boxes (Car Bbox) for ego trajectories across different timesteps.

Autonomous Driving Planning

Autonomous driving planning selects safe, feasible vehicle actions and trajectories amid traffic, uncertainty, road rules, and changing goals.

See all 384 papers

FeasibleFlow: One-Step Joint Transport of Configuration Feasibility and Trajectories for End-to-End Driving
Sep 20, 2026Xiang Li, Bikun Wang, Qing Xu +1Flow Matching Trajectory Generation
Cross-Domain Computing Solutions, Bosch
Vault: One-Step Latent Generation with Positive-Anchored Rewards for Autonomous Driving
Jun 4, 2026Yining Xing, Zehong Ke, Zhiyuan Liu +1End-to-End Autonomous Driving One-Step Generative Modeling
ManeuverFlow: Maneuver Hypothesis Generation with Future Evidence for Trajectory Evaluation and Refinement in End-to-End Autonomous Driving
Oct 9, 2026Yuhao Wei, Yong Chen, Chen Yang +7Autonomous Driving Planning Trajectory Generation
Chongqing University · Artificial Intelligence Center, Geely Automotive Research Institute

Sep 20, 2026J/K move · Enter open · S save

FeasibleFlow: One-Step Joint Transport of Configuration Feasibility and Trajectories for End-to-End Driving

Xiang Li, Bikun Wang, Qing Xu, Jianjun Wang

Cross-Domain Computing Solutions, Bosch

Flow Matching Trajectory Generation End-to-End Autonomous Driving Safe Motion Planning Autonomous Driving Planning

Open paper AskPDF arXiv

End-to-end autonomous driving maps current observations directly to future trajectories, yet those trajectories must remain valid as the scene evolves. Future state modeling aims to address this temporal mismatch, but general representations often contain information unrelated to ego planning and affect trajectory generation only through auxiliary supervision, static conditioning, or proposal evaluation. We propose FeasibleFlow, a one-step end-to-end generative framework that jointly transports a configuration-space feasibility field and multimodal ego trajectories. Our Asymmetric Joint MeanFlow uses the pathwise Jacobian-vector product in the MeanFlow identity to incorporate field evolution into trajectory transport. Because safety feedback is sparser than progress feedback, we further introduce the Anchor-relative ranker (ARR) and Pareto-ReinFlow to balance safety and progress in candidate selection and generation, respectively. Experiments on the NAVSIM benchmark demonstrate the strong performance of FeasibleFlow and validate both the joint transport of feasibility and trajectories and the proposed safety-first mechanisms.

Abstract

Autonomous Driving Planning

Explore similar work

FeasibleFlow: One-Step Joint Transport of Configuration Feasibility and Trajectories for End-to-End Driving

Vault: One-Step Latent Generation with Positive-Anchored Rewards for Autonomous Driving

ManeuverFlow: Maneuver Hypothesis Generation with Future Evidence for Trajectory Evaluation and Refinement in End-to-End Autonomous Driving