Welcome to LocoMuJoCo!
LocoMuJoCo is an imitation learning benchmark specifically targeted towards locomotion. It encompasses a diverse set of environments, including quadrupeds, bipeds, and musculoskeletal human models, each accompanied by comprehensive datasets, such as real noisy motion capture data, ground truth expert data, and ground truth sub-optimal data, enabling evaluation across a spectrum of difficulty levels.
LocoMuJoCo also allows you to specify your own reward function to use this benchmark for pure reinforcement learning! Checkout the example below!
The core idea behind LocoMuJoCo is to allow researcher focussing on imitation or reinforce learning to transition from simple toy task in locomotion to realistic and complex environments crucial for real-world applications. At the same time, we wanted to LocoMuJoCo to be as simple to use as possible by providing comprehensive datasets for each environment and task in a single line of code!