Trajectory Prediction in Autonomous Driving with a Lane Heading Auxiliary Loss

TL;DR

This paper introduces a new differentiable auxiliary loss function for trajectory prediction models in autonomous driving, improving lane-following accuracy and conforming to driving rules, demonstrated on the nuScenes benchmark.

Contribution

It proposes a novel auxiliary loss based on lane heading that enhances multimodal trajectory prediction models by enforcing driving rules.

Findings

Improved off-road and lane violation metrics on nuScenes.

Enhanced model conformity to lane directions.

Better handling of multimodal trajectory predictions.

Abstract

Predicting a vehicle's trajectory is an essential ability for autonomous vehicles navigating through complex urban traffic scenes. Bird's-eye-view roadmap information provides valuable information for making trajectory predictions, and while state-of-the-art models extract this information via image convolution, auxiliary loss functions can augment patterns inferred from deep learning by further encoding common knowledge of social and legal driving behaviors. Since human driving behavior is inherently multimodal, models which allow for multimodal output tend to outperform single-prediction models on standard metrics. We propose a loss function which enhances such models by enforcing expected driving rules on all predicted modes. Our contribution to trajectory prediction is twofold; we propose a new metric which addresses failure cases of the off-road rate metric by penalizing trajectories that oppose the ascribed heading (flow direction) of a driving lane, and we show this metric to be differentiable and therefore suitable as an auxiliary loss function. We then use this auxiliary loss to extend the the standard multiple trajectory prediction (MTP) and MultiPath models, achieving improved results on the nuScenes prediction benchmark by predicting trajectories which better conform to the lane-following rules of the road.