Pedestrian Trajectory Prediction Using Dynamics-based Deep Learning

TL;DR

This paper introduces a dynamics-based deep learning framework that combines a stable dynamical system with a Transformer model to improve pedestrian trajectory prediction, offering better explainability and accuracy.

Contribution

The authors propose a novel integration of an asymptotically stable dynamical system with a Transformer model for pedestrian motion prediction, enhancing interpretability and performance.

Findings

Outperforms existing models on five benchmark datasets.

Provides explicit constraints and explainability in trajectory prediction.

Models human goal-targeted motion effectively.

Abstract

Pedestrian trajectory prediction plays an important role in autonomous driving systems and robotics. Recent work utilizing prominent deep learning models for pedestrian motion prediction makes limited a priori assumptions about human movements, resulting in a lack of explainability and explicit constraints enforced on predicted trajectories. We present a dynamics-based deep learning framework with a novel asymptotically stable dynamical system integrated into a Transformer-based model. We use an asymptotically stable dynamical system to model human goal-targeted motion by enforcing the human walking trajectory, which converges to a predicted goal position, and to provide the Transformer model with prior knowledge and explainability. Our framework features the Transformer model that works with a goal estimator and dynamical system to learn features from pedestrian motion history. The results show that our framework outperforms prominent models using five benchmark human motion datasets.