PACE Dataset for Sim-to-Real Transfer in Legged Robots

This dataset accompanies the PACE framework for actuator model identification in legged robots. It contains raw and processed measurements from actuator drives, single-joint and single-leg experiments, and full-robot locomotion tests across multiple platforms. Python scripts for data fitting and plot generation are included, together with structured NumPy arrays. The dataset supports reproducibility of the experimental results presented in the paper and enables further research on joint dynamics identification and controller transfer.