MIMIC-MJX: Neuromechanical Emulation of Animal Behavior
This provides a tool for neuroscience researchers to analyze neural control strategies and simulate behavioral experiments, though it is incremental as it builds on existing simulation and learning methods.
The authors tackled the problem of inferring neural control policies from kinematic data by introducing MIMIC-MJX, a framework that trains neural controllers to actuate biomechanically-realistic body models in physics simulation to reproduce real kinematic trajectories, resulting in an implementation that is accurate, fast, data-efficient, and generalizable to diverse animal body models.
The primary output of the nervous system is movement and behavior. While recent advances have democratized pose tracking during complex behavior, kinematic trajectories alone provide only indirect access to the underlying control processes. Here we present MIMIC-MJX, a framework for learning biologically-plausible neural control policies from kinematics. MIMIC-MJX models the generative process of motor control by training neural controllers that learn to actuate biomechanically-realistic body models in physics simulation to reproduce real kinematic trajectories. We demonstrate that our implementation is accurate, fast, data-efficient, and generalizable to diverse animal body models. Policies trained with MIMIC-MJX can be utilized to both analyze neural control strategies and simulate behavioral experiments, illustrating its potential as an integrative modeling framework for neuroscience.