We contribute $\text{Bi}^3$, a dataset of social robot navigation among groups of people in a constrained lab space. Compared to prior data collection efforts for social robot navigation, our dataset is unique in that it features: an original experiment design giving rise to close navigation encounters between two humans and a robot; five different navigation algorithms; two different robot platforms; a diverse participant pool of 74 people recruited from two sites in the USA and France; multimodal data streams including 10.5 hours of human and robot ground-truth motion tracks, RGB video, and user impressions over robot performance. Our analysis of the collected crowd behavior through metrics like interaction density and human velocity suggests that $\text{Bi}^3$ represents a benchmark of unique diversity and modeling complexity. Thus, $\text{Bi}^3$ contributes towards understanding how humans and robots can productively mesh their activities in constrained environments, and can be a resource for training models of human motion prediction and robot control policies for navigation in densely crowded spaces.
@inproceedings{stratton2026bi3dataset,
author = {Stratton, Andrew and Singamaneni, Phani Teja and Goyal, Pranav and Alami, Rachid and Mavrogiannis, Christoforos},
title = {Bi3: A Biplatform, Bicultural, Biperson Dataset for Social Robot Navigation},
booktitle = {Proceedings of the IEEE International Conference on Robotics and Automation (ICRA)},
year = {2026},
}We have also published a separate, detailed work on the human subject study this dataset was created in service of. Check it out here.