MFBO_KTH

Learning-enhanced Optimal Gait Design for a Tendon-driven Soft Quadruped Robot via Multi-fidelity Bayesian Optimization

Connect with me:

• 📫kaiget@kth.se
• Or 📫xuezhin@kth.se

Languages and Tools:

Structure

The repository is structured as follows:

• Simulink_models: This directory contains all Simulink files, including soft robot plant models and CPG oscillators.

• data: This directory contains generated data files for 1) offline training results, 2) benchmark method testing, and 3) prerequest data for online training.

• mtgp-master: This directory contains all of the functions used in the MTBO.

• unit_test: This directory contains the main functions for the benchmark method testing and MFBO physical training Python code.

• MFGP_main.m: This m function file is for MFBO physical training.

• CPG_online.slx: This .slx files contains simulink model used for SoftQ online training.

Getting Started

To get started with this project, clone the repository to your local machine using the following command:

git clone https://github.com/n7729697/MTBO_KTH.git