A Knowledge-Driven Diffusion Policy for End-to-End Autonomous Driving Based on Expert Routing

• URL: http://arxiv.org/abs/2509.04853v2
• Date: Mon, 06 Oct 2025 01:23:19 GMT
• Title: A Knowledge-Driven Diffusion Policy for End-to-End Autonomous Driving Based on Expert Routing
• Abstract summary: This paper presents KDP, a knowledge-driven diffusion policy that integrates generative diffusion modeling with a sparse mixture-of-experts routing mechanism.<n>Experiments across representative driving scenarios demonstrate that KDP achieves consistently higher success rates, reduced collision risk, and smoother control compared to prevailing paradigms.<n>These results establish diffusion with expert routing as a scalable and interpretable paradigm for knowledge-driven end-to-end autonomous driving.
• Score: 43.930171390617055
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: End-to-end autonomous driving remains constrained by the difficulty of producing adaptive, robust, and interpretable decision-making across diverse scenarios. Existing methods often collapse diverse driving behaviors, lack long-horizon consistency, or require task-specific engineering that limits generalization. This paper presents KDP, a knowledge-driven diffusion policy that integrates generative diffusion modeling with a sparse mixture-of-experts routing mechanism. The diffusion component generates temporally coherent action sequences, while the expert routing mechanism activates specialized and reusable experts according to context, enabling modular knowledge composition. Extensive experiments across representative driving scenarios demonstrate that KDP achieves consistently higher success rates, reduced collision risk, and smoother control compared to prevailing paradigms. Ablation studies highlight the effectiveness of sparse expert activation and the Transformer backbone, and activation analyses reveal structured specialization and cross-scenario reuse of experts. These results establish diffusion with expert routing as a scalable and interpretable paradigm for knowledge-driven end-to-end autonomous driving.

Related papers

• A Collaborative Multi-Modality Interaction for VLA-based End-to-End Autonomous Driving [96.81243316566764]
  Vision-Language-Action (VLA) models have emerged as a powerful paradigm for end-to-end autonomous driving.<n>We propose a robust VLA-based end-to-end autonomous driving system that combines multi-modality interaction with multi-trajectory planning and optimization.
  arXiv  Detail & Related papers  (2026-08-21T09:06:16Z)
• CoRDE: Concept-Prior Routed Diffusion Experts for Structural Generalization in Robot Manipulation [13.845777652673767]
  Concept-prior Routed Diffusion Experts (CoRDE) is a structure-guided variational distillation framework.<n>CoRDE extracts semantic distributions from a frozen concept encoder to guide the variational posterior responsibility.<n>It forms robust, semantically aligned expert allocations while achieving superior action quality and incremental learning efficiency.
  arXiv  Detail & Related papers  (2026-06-20T08:07:59Z)
• Guiding Neuro-Symbolic Scenario Generation with Spatio-Temporal Logic [3.071933369858584]
  We introduce STRELGen, a framework for the targeted generation of safety-critical driving scenarios.<n>Monitoring satisfaction levels of these specifications is differentiable, enabling gradient-based search.<n> STRELGen thus provides a flexible, interpretable, and powerful tool for stress-testing autonomous driving systems.
  arXiv  Detail & Related papers  (2026-05-18T19:00:14Z)
• Risk-Aware World Model Predictive Control for Generalizable End-to-End Autonomous Driving [82.69496624372944]
  "Only driving like the expert" suffers from limited generalization.<n>Can an E2E-AD system make reliable decisions without any expert action supervision?<n>We propose a unified framework named Risk-aware World Model Predictive Control.
  arXiv  Detail & Related papers  (2026-02-26T17:32:30Z)
• SAME: Stabilized Mixture-of-Experts for Multimodal Continual Instruction Tuning [83.66308307152808]
  We propose StAbilized Mixture-of-Experts (SAME) for Multimodal Continual Instruction Tuning (MCIT)<n>SAME stabilizes expert selection by decomposing routing dynamics into subspaces and updating only task-relevant directions.<n>It also introduces adaptive expert activation to freeze selected experts during training, reducing redundant and cross-task interference.
  arXiv  Detail & Related papers  (2026-02-02T11:47:06Z)
• Large Multimodal Models for Embodied Intelligent Driving: The Next Frontier in Self-Driving? [68.82027978227008]
  This article introduces a novel semantics and policy dual-driven hybrid decision framework to tackle this challenge.<n>The framework merges LMMs for semantic understanding and cognitive representation, and deep reinforcement learning (DRL) for real-time policy optimization.<n>Case study is conducted experimentally to validate the performance superiority of our framework in completing lane-change planning task.
  arXiv  Detail & Related papers  (2026-01-13T11:05:12Z)
• ExpertAD: Enhancing Autonomous Driving Systems with Mixture of Experts [13.846892166111646]
  ExpertAD is a novel framework that enhances the performance of autonomous driving systems.<n>Our experiments show that ExpertAD reduces average collision rates by up to 20% and inference latency by 25% compared to prior methods.
  arXiv  Detail & Related papers  (2025-11-13T13:14:17Z)
• Going Beyond Expert Performance via Deep Implicit Imitation Reinforcement Learning [3.691573844585973]
  We introduce a deep implicit imitation reinforcement learning framework that combines deep reinforcement learning with implicit imitation learning from observation-only datasets.<n>Our main algorithm, Deep Implicit Q-Network (DIIQN), employs an action inference mechanism that reconstructs expert actions through online exploration.<n>We further extend our framework with a Heterogeneous Actions DIIQN (HA-DIIQN) algorithm to tackle scenarios where expert and agent possess different action sets.
  arXiv  Detail & Related papers  (2025-11-05T16:33:39Z)
• Drift-aware Collaborative Assistance Mixture of Experts for Heterogeneous Multistream Learning [31.877595633244734]
  Learning from multiple data streams in real-world scenarios is fundamentally challenging due to intrinsic heterogeneity and unpredictable concept drifts.<n>Existing methods typically assume homogeneous streams and employ static architectures with indiscriminate knowledge fusion.<n>We propose CAMEL, a framework that assigns each stream an independent system with a dedicated feature extractor and task-specific head.<n>Furthermore, we propose an Autonomous Expert Tuner (AET) strategy, which dynamically manages expert lifecycles in response to drift.
  arXiv  Detail & Related papers  (2025-08-03T05:35:34Z)
• Breaking Imitation Bottlenecks: Reinforced Diffusion Powers Diverse Trajectory Generation [20.106116218594266]
  DIVER is an end-to-end autonomous driving framework that integrates reinforcement learning and diffusion-based generation.<n>We show that DIVER significantly improves trajectory diversity, effectively addressing the mode collapse problem inherent in imitation learning.
  arXiv  Detail & Related papers  (2025-07-05T14:19:19Z)
• Rethinking Explainability in the Era of Multimodal AI [9.57008593971486]
  multimodal AI systems have become ubiquitous and achieved remarkable performance across high-stakes applications.<n>Most existing explainability techniques remain unimodal, generating modality-specific feature attributions, concepts, or circuit traces in isolation.<n>This paper argues that such unimodal explanations systematically misrepresent and fail to capture the cross-modal influence that drives multimodal model decisions.
  arXiv  Detail & Related papers  (2025-06-16T03:08:29Z)
• TeLL-Drive: Enhancing Autonomous Driving with Teacher LLM-Guided Deep Reinforcement Learning [61.33599727106222]
  TeLL-Drive is a hybrid framework that integrates a Teacher LLM to guide an attention-based Student DRL policy.<n>A self-attention mechanism then fuses these strategies with the DRL agent's exploration, accelerating policy convergence and boosting robustness.
  arXiv  Detail & Related papers  (2025-02-03T14:22:03Z)
• Planning-Aware Diffusion Networks for Enhanced Motion Forecasting in Autonomous Driving [0.0]
  Planning-Integrated Forecasting Model (PIFM) is a novel framework inspired by neural mechanisms governing decision-making and multi-agent coordination in the brain. PIFM is able to forecast future trajectories of all agents within a scenario. This architecture enhances model transparency, as it parallels the brain's method of dynamically adjusting predictions based on external stimuli and other agents'behaviors.
  arXiv  Detail & Related papers  (2024-10-25T15:44:51Z)
• DeepInteraction++: Multi-Modality Interaction for Autonomous Driving [80.8837864849534]
  We introduce a novel modality interaction strategy that allows individual per-modality representations to be learned and maintained throughout.<n>DeepInteraction++ is a multi-modal interaction framework characterized by a multi-modal representational interaction encoder and a multi-modal predictive interaction decoder.<n>Experiments demonstrate the superior performance of the proposed framework on both 3D object detection and end-to-end autonomous driving tasks.
  arXiv  Detail & Related papers  (2024-08-09T14:04:21Z)
• Collision Probability Distribution Estimation via Temporal Difference Learning [0.46085106405479537]
  We introduce CollisionPro, a pioneering framework designed to estimate cumulative collision probability distributions. We formulate our framework within the context of reinforcement learning to pave the way for safety-aware agents. A comprehensive examination of our framework is conducted using a realistic autonomous driving simulator.
  arXiv  Detail & Related papers  (2024-07-29T13:32:42Z)
• Traj-MAE: Masked Autoencoders for Trajectory Prediction [69.7885837428344]
  Trajectory prediction has been a crucial task in building a reliable autonomous driving system by anticipating possible dangers. We propose an efficient masked autoencoder for trajectory prediction (Traj-MAE) that better represents the complicated behaviors of agents in the driving environment. Our experimental results in both multi-agent and single-agent settings demonstrate that Traj-MAE achieves competitive results with state-of-the-art methods.
  arXiv  Detail & Related papers  (2023-03-12T16:23:27Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.