Data: We generate and curate high-resolution datasets on collective behaviour across species and contexts, combining field, lab, and applied systems. Most data are released with publications; some legacy datasets are shared on request.

​

Shepherding dataset: One of our most requested datasets (almost weekly!) captures collective motion under external control. It comprises three herding events involving 46 female Merino sheep and a trained sheepdog, all tracked at 1 Hz using custom GPS loggers with 10–20 cm positional accuracy (designed by the RVC Structure & Motion Lab). These data have been used to investigate selfish-herd behaviour of sheep and build a model for shepherding.

​​

Data are shown on the right: (A) flock cohesion as a function of dog-distance for three herding trials, and (B) individual sheep movements relative to the flock centroid.​​​​​ ​Download the data for each trial here. 

​

When using the data, please cite our two papers [1],[2], and send me an email to let me know how you get on/what you use the data for! ​

​