Learning to Predict Vehicle Trajectories with Model-based Planning

Abstract

Predicting the future trajectories of on-road vehicles is critical for autonomous driving. In this paper, we introduce a novel prediction framework called PRIME, which stands for Prediction with Model-based Planning. Unlike recent prediction works that utilize neural networks to model scene context and produce unconstrained trajectories, PRIME is designed to generate accurate and feasibility-guaranteed future trajectory predictions. PRIME guarantees the trajectory feasibility by exploiting a model-based generator to produce future trajectories under explicit constraints and enables accurate multimodal prediction by utilizing a learning-based evaluator to select future trajectories. We conduct experiments on the large-scale Argoverse Motion Forecasting Benchmark, where PRIME outperforms the state-of-the-art methods in prediction accuracy, feasibility, and robustness under imperfect tracking.

arXiv commentAccepted by The Conference on Robot Learning (CoRL) 2021. Project page at http://haoran-song.github.io/prime

References in corpus (5)

• Social Force Model for Pedestrian Dynamics
• IntentNet: Learning to Predict Intention from Raw Sensor Data
• TNT: Target-driveN Trajectory Prediction
• Argoverse: 3D Tracking and Forecasting with Rich Maps
• Multiple Futures Prediction

Cited by in corpus (6)

• Social Interactions for Autonomous Driving: A Review and Perspectives
• THOMAS: Trajectory Heatmap Output with learned Multi-Agent Sampling
• Trajectory Prediction for Autonomous Driving: Progress, Limitations, and Future Directions
• DenseTNT: End-to-end Trajectory Prediction from Dense Goal Sets
• GIN: Graph-based Interaction-aware Constraint Policy Optimization for Autonomous Driving
• Trajectory Prediction with Graph-based Dual-scale Context Fusion