Trajectory Forecasts in Unknown Environments Conditioned on Grid-Based Plans

TL;DR

This paper introduces a novel approach for trajectory forecasting in unknown environments by conditioning on grid-based plans learned through maximum entropy inverse reinforcement learning, resulting in diverse and scene-conforming predictions.

Contribution

It reformulates MaxEnt IRL to jointly infer agent goals and paths on a grid, and proposes an attention-based generator for continuous trajectory prediction.

Findings

Model produces diverse multimodal trajectory predictions.

Predictions conform well to scene structure over long horizons.

Outperforms existing methods on Stanford drone and NuScenes datasets.

Abstract

We address the problem of forecasting pedestrian and vehicle trajectories in unknown environments, conditioned on their past motion and scene structure. Trajectory forecasting is a challenging problem due to the large variation in scene structure and the multimodal distribution of future trajectories. Unlike prior approaches that directly learn one-to-many mappings from observed context to multiple future trajectories, we propose to condition trajectory forecasts on plans sampled from a grid based policy learned using maximum entropy inverse reinforcement learning (MaxEnt IRL). We reformulate MaxEnt IRL to allow the policy to jointly infer plausible agent goals, and paths to those goals on a coarse 2-D grid defined over the scene. We propose an attention based trajectory generator that generates continuous valued future trajectories conditioned on state sequences sampled from the MaxEnt policy. Quantitative and qualitative evaluation on the publicly available Stanford drone and NuScenes datasets shows that our model generates trajectories that are diverse, representing the multimodal predictive distribution, and precise, conforming to the underlying scene structure over long prediction horizons.