Foresight in Motion: Reinforcing Trajectory Prediction with Reward Heuristics

TL;DR

This paper introduces a planning-based trajectory prediction method for autonomous driving that uses reward heuristics and intention reasoning to improve accuracy and confidence in motion forecasting.

Contribution

It proposes a novel query-centric IRL scheme for intention reasoning and integrates it with a hierarchical decoder for improved trajectory prediction.

Findings

Significantly improves prediction confidence on Argoverse and nuScenes datasets.

Achieves state-of-the-art or competitive performance in motion forecasting.

Demonstrates the effectiveness of reward heuristics in trajectory prediction.

Abstract

Motion forecasting for on-road traffic agents presents both a significant challenge and a critical necessity for ensuring safety in autonomous driving systems. In contrast to most existing data-driven approaches that directly predict future trajectories, we rethink this task from a planning perspective, advocating a "First Reasoning, Then Forecasting" strategy that explicitly incorporates behavior intentions as spatial guidance for trajectory prediction. To achieve this, we introduce an interpretable, reward-driven intention reasoner grounded in a novel query-centric Inverse Reinforcement Learning (IRL) scheme. Our method first encodes traffic agents and scene elements into a unified vectorized representation, then aggregates contextual features through a query-centric paradigm. This enables the derivation of a reward distribution, a compact yet informative representation of the target agent's behavior within the given scene context via IRL. Guided by this reward heuristic, we perform policy rollouts to reason about multiple plausible intentions, providing valuable priors for subsequent trajectory generation. Finally, we develop a hierarchical DETR-like decoder integrated with bidirectional selective state space models to produce accurate future trajectories along with their associated probabilities. Extensive experiments on the large-scale Argoverse and nuScenes motion forecasting datasets demonstrate that our approach significantly enhances trajectory prediction confidence, achieving highly competitive performance relative to state-of-the-art methods.