ExecDiffuser improves decision making for self driving vehicles
ExecDiffuser: Closing the Sample-to-Execution Gap in Diffusion Planning
Robotics
Summary
Planning paths for self-driving cars often generates many possible routes, but only one path is chosen to follow. The authors found that making more candidate paths doesn’t always lead to better driving decisions. They created ExecDiffuser, a system that connects path creation, selection, and replanning so the chosen route better matches what the car actually drives. Tests show ExecDiffuser chooses better driving commands, improving performance on several driving simulations.
What this means in practice
- •For autonomous vehicle engineers: Improve real-time path planning systems by integrating candidate generation and selection for safer control execution in self-driving cars.
- •For simulation platform developers: Enhance driving simulators with planning methods that better match simulated executed trajectories to candidate plans for more realistic scenario testing.
Authors
Shuo Pei, Bosun Liang, Saihui Zha, Chen Sun, Huachun Tan, Yong Wang
Abstract
Generative planners are often evaluated by the trajectories they can generate, although closed-loop driving ultimately depends on the single candidate committed to control. We identify this mismatch as the sample-to-execution gap and show that improving proposal quality alone can worsen the executed decision. Here, we introduce ExecDiffuser, a recurrent proposal-to-execution framework that couples candidate generation, trajectory selection, and subsequent replanning through shared execution semantics. Intent-Guided Diffusion Proposal (IGD) assigns hypothesis responsibility and shapes trajectories with execution-conditioned supervision. Execution-Grounded Selection (EGS) learns a candidate-conditioned publication utility, commits one trajectory to control, and returns it as context for the next planning cycle. Across nuPlan, InterPlan, and NAVSIM, ExecDiffuser achieves leading closed-loop performance against strong state-of-the-art baselines. Under a unified five-seed protocol, it obtains an average score of 85.62, improving over Diffusion Planner and FlowDrive by 4.02 and 1.31 points, respectively. On NAVSIM, it achieves 92.6 PDMS under the official evaluation pipeline. Matched counterfactual analysis further shows that ExecDiffuser more reliably converts strong candidates into committed decisions, while prefix interventions confirm that this advantage persists after physical execution and subsequent replanning.