Flash: Fast and Light Motion Prediction for Autonomous Driving with Bayesian Inverse Planning and Learned Motion Profiles

TL;DR

This paper introduces a modular, Bayesian inverse planning-based motion prediction system for autonomous driving that combines goal extraction, trajectory generation, and neural motion profiles, achieving state-of-the-art results and interpretability.

Contribution

The novel integration of Bayesian inverse planning with lightweight neural networks for motion profile prediction enhances interpretability and performance in autonomous driving scenarios.

Findings

Achieves state-of-the-art trajectory error on NGSIM dataset.

Provides detailed analysis of system components and behavior stratification.

Demonstrates interpretability and physical feasibility of predicted trajectories.

Abstract

Motion prediction of road users in traffic scenes is critical for autonomous driving systems that must take safe and robust decisions in complex dynamic environments. We present a novel motion prediction system for autonomous driving. Our system is based on the Bayesian inverse planning framework, which efficiently orchestrates map-based goal extraction, a classical control-based trajectory generator and a mixture of experts collection of light-weight neural networks specialised in motion profile prediction. In contrast to many alternative methods, this modularity helps isolate performance factors and better interpret results, without compromising performance. This system addresses multiple aspects of interest, namely multi-modality, motion profile uncertainty and trajectory physical feasibility. We report on several experiments with the popular highway dataset NGSIM, demonstrating state-of-the-art performance in terms of trajectory error. We also perform a detailed analysis of our system's components, along with experiments that stratify the data based on behaviours, such as change-lane versus follow-lane, to provide insights into the challenges in this domain. Finally, we present a qualitative analysis to show other benefits of our approach, such as the ability to interpret the outputs.