Documentation

In SCONE, everything needed to perform a predictive simulation is bundled in what is called a .scone scenario file, which contains:

• A Model of the human, animal or robot you wish to simulate. Make sure your model is suitable for predictive simulations.
• A Controller that produces input signals for the model actuators. Controllers can be open-loop or closed-loop (or a combination).
• An Objective that describes the goal task for which you wish to optimize, through a weighted combination of Measures.
• An Optimizer that attempts to find the model and control parameters that optimally perform the specified objective.

To learn more about SCONE scenarios, be sure check out the SCONE Tutorials, or see the following topics:

• Parameter Initialization
• Batch Optimization
• Mouse and Keyboard Shortcuts
• The .sto File Format
• The .par File Format

SconePy is a new Python API designed for RL applications. See the SconePy API Reference for more details.