On Improving Multimodal Pedestrian Trajectory Prediction with CVAE: A Study on Benchmark and Robot Data

TL;DR

This paper enhances pedestrian trajectory prediction by integrating a CVAE with Social-STGCNN, improving diversity and accuracy of predictions on benchmark and real-world robot data.

Contribution

It introduces a CVAE-based probabilistic model built on Social-STGCNN to better capture multimodal pedestrian trajectories.

Findings

Moderate improvements on ETH and UCY datasets.

More consistent endpoint accuracy and trajectory diversity.

Effective performance on real-world robot-collected data.

Abstract

Accurate pedestrian trajectory prediction is crucial for autonomous systems operating in complex environments, such as modular buses and delivery robots in suburban or semi-structured areas. Social Spatio-Temporal Graph Convolutional Neural Networks (Social-STGCNN) have shown strong performance by modeling social interactions; however, producing diverse and well-calibrated future trajectories remains challenging. In this work, we build on a Social-STGCNN backbone and introduce a Conditional Variational Autoencoder (CVAE)-based probabilistic formulation to explicitly model multimodal future trajectories. We evaluate the method on the ETH and UCY pedestrian trajectory datasets as well as on a real-world pedestrian dataset collected by a mobile robot. Results show moderate gains on public benchmarks, but more consistent endpoint accuracy and improved trajectory diversity across different crowd configurations. Evaluation on robot-collected data further demonstrates the approach's effectiveness beyond curated benchmarks and supports its applicability in practical deployments.