Tutorials used in the Ghent University course: Software Engineering Lab 3 (SEL3).
The general goal of the course's project is to optimize controllers for an in-silico brittle star robot. The simulation model and environment are provided by the Bio-inspired Robotics Testbed (BRT).
Tutorials:
- brb_brittle_star_environment.ipynb: covers the creation and usage of the BRT's brittle star simulation environment.
- brb_brittle_star_cpg_tutorial.ipynb: covers the background and implementation of Central Pattern Generators. Provides a case study in which a CPG model is applied to generate a simplified rowing gait for the brittle star robot.
- brb_brittle_star_qlearning_tutorial.ipynb: covers the implementation and application of Q-Learning (a canonical reinforcement learning algorithm). Provides a case study in which Q-Learning is applied to optimize a controller for the brittle star.
- brb_brittle_star_evolution_strategy_tutorial.ipynb: using an evolution strategy to optimise the modulation parameters of a CPG for brittle star locomotion.