SDPAD: A Fully Spike-Driven Pipeline for End-to-End Autonomous Driving

Chengjun Zhang; Yuhao Zhang; Jie Yang; Mohamad Sawan

cs.ROOct 8, 2026

SDPAD: A Fully Spike-Driven Pipeline for End-to-End Autonomous Driving

Authors: Chengjun Zhang, Yuhao Zhang, Jie Yang, Mohamad Sawan

Organizations: Zhejiang Key Laboratory of 3D Micro/Nano Fabrication and Characterization, Westlake Institute for Optoelectronics, Fuyang, Hangzhou, China · Integrated-On-Chips Brain-Computer Interfaces Zhejiang Engineering Research Center, Hangzhou, Zhejiang, China · CenBRAIN Neurotech, School of Engineering, Westlake University, Hangzhou, Zhejiang, China

Abstract

End-to-end autonomous driving demands trajectory planners that are both highly accurate and cheap enough for edge deployment. State-of-the-art artificial neural network (ANN) planners meet the accuracy requirement at the cost of heavy dense computation, while spiking neural networks (SNNs)---though promising orders-of-magnitude energy savings through sparse, event-driven arithmetic---still lag far behind in planning accuracy. We present \textbf{SDPAD}, a fully spike-driven end-to-end planning pipeline that closes this gap. SDPAD converts a pre-trained ANN perception stack into integer-spike form via quantized ANN2SNN conversion, lifts multi-view images into the bird's-eye-view (BEV) space with a spike-driven-max (SDM) depth distribution (Spike-3D-Lift), and plans through the Spike-QFormer, a spiking query transformer in which ego, agent, and map queries distilled from the BEV scene are fused by learnable waypoint queries via cross-attention, followed by deformable spike-cross-attention refinement. Every operation is gated by integer spikes and inference is a single feed-forward pass without temporal simulation loops. On the nuScenes open-loop benchmark, SDPAD achieves an average $L_2$ error of 0.40,m and a collision rate of 0.12%, on par with strong ANN planners while consuming 69.9,mJ---less than 2% of recent ANN baselines. In closed-loop evaluation on the NAVSIM navtest split, SDPAD reaches 86.3 PDMS, surpassing the previous SNN planner SAD by 4.3 points and matching mainstream ANN planners at a fraction of their energy. To our knowledge, SDPAD is the first fully spike-driven planner evaluated in end-to-end autonomous driving, demonstrating that SNNs can rival dense ANNs in complex driving tasks.

Figures & tables

Figure 1: The comparison of different end-to-end paradigms. (a) Dense ANN pipelines compute in floating point throughout. (b) Existing SNN planners still rely on continuous membrane potentials or softmax depth scores for BEV lifting. (c) SDPAD is fully spike-driven end to end: the proposed Spike-3D-Lift turn BEV sampling into spike BEV representation.

Figure 2: Overall architecture of the proposed Pipline. Multi - view images are encoded into a Spiking BEV representation via E - Spikeformer and Spike - 3D - Lift, which simultaneously supports semantic segmentation and 3D object detection. The planning stage then refines waypoints with a 3 - mode reference curve and Deformable SCA, and finally outputs the ego - vehicle trajectory.

Figure 3: The illustration of the Spike - 3D-Lift for Spike-BEV generation process. Image features are weighted by the accumulated spiking depth scores to form 3D frustum features (membrane potentials), followed by camera - based interpolation and spiking neural conversion to Spiking BEV Representation.

Method	SNN	Img. Backbone	Auxiliary Task	L2 (m) $\downarrow$				Collision Rate (%) $\downarrow$				Energy (mJ) $\downarrow$
Method	SNN			1s	2s	3s	Avg.	1s	2s	3s	Avg.	Energy (mJ) $\downarrow$
ST-P3 Hu et al. (2022)	✗	EffNet-b4	Det& Map	1.33	2.11	2.90	2.11	0.23	0.62	1.27	0.71	3520.40
UniAD Hu et al. (2023)	✗	ResNet-101	Det&Track&Map&Motion&Occ	0.45	0.70	1.04	0.73	0.62	0.58	0.63	0.61	-
VAD Jiang et al. (2023)	✗	ResNet-50	Det&Map&Motion	0.41	0.70	1.05	0.72	0.07	0.17	0.41	0.22	-
SparseDrive Sun et al. (2025)	✗	ResNet-50	Det&Track&Map&Motion	0.29	0.58	0.96	0.61	0.01	0.05	0.18	0.08	-
DiffusionDrive Liao et al. (2025)	✗	ResNet-50	Det&Track&Map&Motion	0.27	0.54	0.90	0.57	0.03	0.05	0.16	0.08	-

Table 1: Comparison on nuScenes dataset with open-loop metrics. metric calculated in the same way as VAD

Method	SNN	Img. Backbone	Auxiliary Task	NAC $\uparrow$	DAC $\uparrow$	TTC $\uparrow$	Comf. $\uparrow$	EP $\uparrow$	PDMS $\uparrow$	Energy (mJ) $\downarrow$
Transfuser Chitta et al. (2022)	✗	ResNet-34	Det&Map	97.7	92.8	92.8	100	79.2	84.0	114.46
UniAD Hu et al. (2023)	✗	ResNet-34	Det&Map	97.8	91.9	92.9	100	78.8	83.4	-
VADv2 Chen et al. (2024b)	✗	ResNet-34	Det&Map	97.2	89.1	91.6	100	76.0	80.9	-
Hydra-MDP-W-EP Li et al. (2024)	✗	ResNet-34	Det&Map	98.3	96.0	94.6	100	78.7	86.5	-
LAW Li et al. (2025)	✗	ResNet-34	None	96.4	95.4	88.7	99.9	81.7	84.6	-
World4Drive Zheng et al. (2025)	✗	ResNet-34	None	97.4	94.3	92.8	100	79.9	85.1	-

Table 2: Comparison of state-of-the-art methods on the NAVSIM navtest split.

ID	Ego Query	Agent Query	Map Query	Deformable -SCA	Param. (M) $\downarrow$	L2 (m) $\downarrow$			Collision Rate (%) $\downarrow$			Energy (mJ) $\downarrow$
ID	Ego Query	Agent Query	Map Query	Deformable -SCA	Param. (M) $\downarrow$	1s	2s	3s	1s	2s	3s	Energy (mJ) $\downarrow$
1	✗	✗	✓	✗	34.91	0.28	0.57	0.98	0.04	0.19	0.44	69.27
2	✓	✗	✓	✗	38.11	0.27	0.56	0.97	0.02	0.16	0.51	68.71
3	✗	✓	✓	✗	38.38	0.20	0.39	0.70	0.02	0.12	0.35	69.58
4	✓	✓	✗	✗	35.21	0.21	0.38	0.67	0.02	0.11	0.34	70.02
5	✓	✓	✓	✗	38.39	0.20	0.38	0.66	0.02	0.12	0.30	69.64

Table 3: Ablation for design choices.

Table 7

Appendix figures & tables4 assets

Supplementary material from the paper’s appendix.

Appendix

Module	Specification
Image backbone	Spike-driven transformer (large), 19.0M
	embed dims $[64,128,256,360]$ , stride-16 output
Spike-3D-Lift	Radial–Cartesian sampling Zhang et al. [2025]
	SDM depth neuron, $V_{\max}{=}8$ , BEV channels 80
BEV encoder	channels $[128,256,360,256,256]$ , output 256
	agent-view features $10{\times}10$

Appendix

Table B.6: Per-stage architecture specification of SDPAD.

Method	Backbone	Image Size	Frames	mAP $\uparrow$	NDS $\uparrow$	mATE $\downarrow$	mASE $\downarrow$	mAOE $\downarrow$	mAVE $\downarrow$	mAAE $\downarrow$
BEVDet (arXiv. 2021)	ResNet50	256 $\times$ 704	1	0.298	0.379	0.725	0.279	0.589	0.860	0.245
PETRv2 (ICCV. 2023)	ResNet50	256 $\times$ 704	2	0.349	0.456	0.700	0.275	0.580	0.437	0.187
BEVDepth (AAAI. 2023)	ResNet50	256 $\times$ 704	2	0.351	0.475	0.639	0.267	0.479	0.428	0.198
BEVStereo (AAAI. 2023)	ResNet50	256 $\times$ 704	2	0.372	0.500	0.598	0.270	0.438	0.367	0.190
SA-BEV (ICCV. 2023)	ResNet50	256 $\times$ 704	2	0.387	0.512	0.613	0.266	0.352	0.382	0.199
BEFormerv2 (CVPR. 2023)	ResNet50	-	-	0.423	0.529	0.618	0.273	0.413	0.333	0.188

Appendix

Table C.7: Comparison with previous state-of-the-art multi-view 3D detectors on the nuScenes val set.

Figure D.4: Predicted trajectories of SDPAD versus ground truth in diverse nuScenes scenarios, together with the BEV segmentation and agent detection outputs. We also show the intermediate Spiking BEV features.

Figure D.5: Predicted trajectories of SDPAD versus ground truth in diverse NAVSIM.

Explore similar work

PrismAD: Decoupled Planning via Semantic Mixture-of-Planners for End-to-End Autonomous Driving
Jul 11, 2026Kang Ding, Zhigui Lin, Hongsong Wang +5End-to-End Autonomous Driving Mixture of Experts
School of Vehicle and Mobility, Tsinghua University, Beijing 100084, China. · State Key Laboratory of Intelligent Green Vehicle and Mobility, Tsinghua University, Beijing 100084, China. · School of Cyberspace Security, Southeast University, Nanjing 210096, China. +3
Neuromorphic LiDAR-based Bird's Eye View Object Detection using Energy-efficient Spiking Neural Networks
May 24, 2026Sambit Mohapatra, Senthil Yogamani, Heinrich Gotzig +1Lidar-Based 3D Object Detection Autonomous Driving Perception
Valeo, Germany · TU Ilmenau, Germany · Valeo, Ireland
Drive-JEPA: Video JEPA Meets Multimodal Trajectory Distillation for End-to-End Driving
Jan 29, 2026Linhan Wang, Zichong Yang, Chen Bai +6End-to-End Autonomous Driving Joint-Embedding Predictive Architecture
Virginia Tech · Purdue University · XPENG Motors +1

Jul 11, 2026, cs.ROJ/K move · Enter open · S save

PrismAD: Decoupled Planning via Semantic Mixture-of-Planners for End-to-End Autonomous Driving

Kang Ding, Zhigui Lin, Hongsong Wang, Jie Gui+4

School of Vehicle and Mobility, Tsinghua University, Beijing 100084, China. · State Key Laboratory of Intelligent Green Vehicle and Mobility, Tsinghua University, Beijing 100084, China. · School of Cyberspace Security, Southeast University, Nanjing 210096, China.+3

End-to-End Autonomous Driving Mixture of Experts Autonomous Driving Planning

Open paper AskPDF arXiv

This letter presents PrismAD, a decoupled end-to-end autonomous driving framework based on a Semantic Mixture-of-Planners. Existing planners usually aggregate heterogeneous scene tokens into a coupled representation space, forcing a single planning branch to jointly model agent interaction, road geometry, and driving intention. Such coupling may weaken factor-specific reasoning and obscure the contribution of different planning cues. To address this limitation, PrismAD partitions scene tokens into interaction, geometry, and intent groups, and assigns them to independent planning experts with the same architecture but separate parameters. Each expert learns a specialized motion-planning representation, while a semantics-aware router adaptively aggregates expert predictions with separate routing weights for motion prediction and ego planning. Sparse top-

K

activation with noisy gating is further introduced to improve routing robustness and reduce unnecessary expert computation. Extensive experiments on the nuScenes open-loop dataset and NeuroNCAP closed-loop benchmark demonstrate that PrismAD exhibits competitive performance. Our code will be released soon.

Abstract

Explore similar work

PrismAD: Decoupled Planning via Semantic Mixture-of-Planners for End-to-End Autonomous Driving

Neuromorphic LiDAR-based Bird's Eye View Object Detection using Energy-efficient Spiking Neural Networks

Drive-JEPA: Video JEPA Meets Multimodal Trajectory Distillation for End-to-End Driving