Removing av traffic pairing subtly affects planner route choices
From Learned-Mode AV-Traffic Pairing to Planner Decisions: A Marginal-Preserving Study on Argoverse 2
Robotics
Summary
Predicting how self-driving cars and nearby vehicles will move together is important for planning safe routes. This paper studies what happens if the connection between predicted car and traffic paths is removed while keeping other parts the same. The authors found that removing this pairing changes some route decisions, even though usual prediction accuracy scores stay the same. However, they couldn’t conclusively say whether the pairing itself or related changes caused the decision differences.
What this means in practice
- •For autonomous vehicle engineers: Evaluate how forecasts that link AV and traffic predictions influence route planning decisions in simulation scenarios.
- •For transportation simulation teams: Test how removing correlated traffic forecasts impacts decision models without changing individual trajectory statistics.
Tested on one dataset.
Authors
Jingyu Wang
Abstract
Joint motion forecasts pair each autonomous-vehicle (AV) future with surrounding traffic, but actor-level metrics do not show whether that structure matters to a planner. We study this question with a marginal-preserving product control that removes AV-traffic pairing among the learned modes while retaining the fixed constant-velocity pair and holding trajectories, actor-level marginals before planner conditioning, candidates, the cost terms and weights, and fallback fixed. The intervention also changes candidate-conditioned concentration. Across twelve runs on 1,400 held-out Argoverse 2 scenarios, the intervention changes 3.0% of route-level offline selections at $τ=4$ m. Control-minus-joint recorded-trajectory regret is $-0.026$ and $-0.118$ at the two training sizes; crossed and seed-$t$ intervals span zero. At $τ=1$ m, relative costs change in 87.9% of route evaluations and route-level offline selections in 8.1%. Before concentration matching, descriptive outcome estimates favor the control. Most of this gap disappears along an approximate concentration-matching path; the remaining contrasts are $+0.112$ and $-0.047$, and both crossed intervals span zero. Actor-level forecast metrics remain identical. Pairing-strength and temperature sweeps show that the decision contrast grows with pairing removal and sharper conditioning. The intervention changes planner decisions even though actor-level metrics remain unchanged. The matching analysis, however, cannot separate any recorded-outcome effect of learned-mode AV-traffic pairing from the accompanying change in conditioned concentration.