On Improving Multimodal Pedestrian Trajectory Prediction with CVAE: A Study on Benchmark and Robot Data
For autonomous systems needing diverse and accurate pedestrian predictions, this work offers an incremental improvement by integrating CVAE into an existing graph-based model.
The authors improved multimodal pedestrian trajectory prediction by adding a CVAE to a Social-STGCNN backbone, achieving moderate gains on ETH/UCY benchmarks but more consistent endpoint accuracy and diversity, and showing effectiveness on real robot-collected data.
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.