UniTeD: Unified Temporal Diffusion for Joint Perception and Planning in Autonomous Driving

2026-06-24 · Source: Takara TLDR - Daily AI Papers · Field: Transportation & Mobility — Autonomous Vehicles & Smart Transportation, Robotics & Autonomous Systems, Artificial Intelligence & Machine Learning · Depth: Expert, medium

Summary

UniTeD is a Unified Temporal Diffusion framework designed for joint perception and planning in end-to-end autonomous driving, addressing limitations of existing decoupled systems that propagate perception errors to the planner. This framework models both tasks through iterative denoising within a shared generative space, enabling bidirectional information exchange and mutual refinement. UniTeD enhances robustness via noise-conditioned multi-task training. It extends to streaming scenarios by incorporating a Temporal Transition Module (TTM) to manage noise-level mismatches between frames and introduces an Anchor Refresh Strategy (ARS) to mitigate training-inference distribution shifts common in sparse diffusion-based driving frameworks. UniTeD achieves state-of-the-art performance on multiple benchmarks, outperforming recent discriminative end-to-end and diffusion-based planning methods.

Key takeaway

For Machine Learning Engineers developing autonomous driving systems, UniTeD offers a robust alternative to decoupled perception and planning architectures. You should consider adopting unified temporal diffusion models to mitigate error propagation and enhance system robustness. This approach, incorporating temporal context and strategies like ARS, can lead to state-of-the-art performance and simplify your end-to-end optimization challenges.

Key insights

Unified Temporal Diffusion (UniTeD) jointly models perception and planning, improving robustness and performance in autonomous driving.

Principles

• Bidirectional information exchange refines perception and planning.
• Noise-conditioned multi-task training enhances system robustness.
• Temporal context integration improves streaming performance.

Method

UniTeD employs iterative denoising in a shared generative space, integrates a Temporal Transition Module (TTM) for noise-level matching, and uses an Anchor Refresh Strategy (ARS) to reduce training-inference distribution shift.

In practice

• Apply unified diffusion for end-to-end autonomous driving.
• Implement TTM for temporal context in streaming systems.
• Use ARS to stabilize sparse diffusion model training.

Topics

• Autonomous Driving
• Diffusion Models
• End-to-End Planning
• Perception Systems
• Temporal Modeling
• Machine Learning Engineering

Code references

• xiaomi-research/unidrivevla
• happinesslz/UniLION
• hey-cjj/EvoDriveVLA
• happinesslz/DrivePI

Best for: Research Scientist, AI Scientist, Machine Learning Engineer, Computer Vision Engineer

Related on AIssential

• UniDrive: A Unified Vision-Language and Grounding Framework for Interpretable Risk Understanding in Autonomous Driving — UniDrive unifies temporal reasoning and high-resolution perception for interpretable, spatially precise risk understanding in autonomous driving.
• Proposal-Conditioned Latent Diffusion for Closed-Loop Traffic Scenario Generation — A new diffusion framework improves closed-loop traffic simulation efficiency and control for autonomous vehicles.
• Unified Map Prior Encoder for Mapping and Planning — Unified, alignment-aware fusion of heterogeneous map priors significantly enhances autonomous driving mapping and planning.
• CooperDrive: Enhancing Driving Decisions Through Cooperative Perception — Cooperative perception enhances autonomous vehicle safety by expanding situational awareness and enabling proactive decision-making.
• Towards End to End Motion Planning and Execution for Autonomous Underwater Vehicles Using Reinforcement Learning — An end-to-end hierarchical DRL system enables AUV navigation from raw sensor data to thruster commands.
• PLAN-S: Bridging Planning with Latent Style Dynamics for Autonomous Driving World Models — PLAN-S improves autonomous driving LWMs by using style-conditioned semantic cost maps for explicit risk and style modulation in planning.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Takara TLDR - Daily AI Papers.