top of page
Evolutionary Computation and Design
Evolving softrobots
In this course, we begin exploring the world of genetic algorithms and how they are used to solve real life problems. For this project we begin out by creating a physics simulator from scratch and consequently modelling robots of different shapes. we then train the robots to evolve over time to learn to walk/move over time.
Evolutionary Computation and Design: Project
The first step in this project is to create a physics simulator with gravity, damping and friction, from scratch. And then model a simple cube and make it bounce.
Following which we then made robots with arbitrary configurations of masses and springs. We chose to evolve the gait of a snake like robot as seen in the pictures with the yellow bars.
To configure the gait of the robot, we varied spring parameters like, K (spring constant), omega and Phi (spring frequency and. phase).
we then employed an Evolutionary algorithm to best learn these gait parameters that would maximise moving speed.
Evolutionary Computation and Design: Text
Evolutionary Computation and Design: Pro Gallery
Evolutionary Computation and Design: Video Player
bottom of page