Trajectory Augmentation for Robust Neural Locomotion Controllers

In this work, we show that this is due to training on
unbalanced biased datasets with poor representation for many important gaits and transitions. To solve this
poor data problem, we introduce Trajectory Augmentation (TrajAug), a fully automatic data augmentation
technique that generates synthetic motion data by using motion matching to stitch sequences from the original
dataset to follow random trajectories. By uniformly sampling these trajectories, we can rebalance the dataset
and introduce sharper turns that are commonly used in-game but are hard to capture.
Publication Link: studios.disneyresearch.com/20...