Hi. My name is Travis. This is my GitHub. My work focuses mainly on control and learning in multi-agent robotics applications. I prefer Python for most things.
My primary development project is the multi-agent coordination simulator, where I integrate various swarming techniques into a single environment for use in my academic research. You will see various offshoots where I play with new concepts, but they typically get intergated back into this project.
Here are some things I am working on:
- Latent Linear Manifolds: DNN-based autoencoder to learn the linear essence of swarm dynamics.
- Hypospace Learning: Learning and control of swarms in lower-dimensional hypospace
- Twisted Circles: Forming simple closed curves by twisting circles.
- Finding Neighbours: An algorithm for finding consistent neighbours in busy trajectories.
- In a private archive, I am also working on image processing on the Adver-City dataset.
- Various other tangents and reflections found on personal website here.
