D'Kitty
D'Kitty (pronounced as "The Kitty") is a twelve degree of freedom platform for exploring learning-based techniques in locomotion.
Hardware
Simulation
Schematics
Gallery
Gallery
Indoor: Walking in clutter
A Sim2Real policy trained via Natural Policy Gradient on MuJoCo simulation with randomized perturbations learns to walk in clutter and step over objects
Indoor: Walking through obstacles
A Sim2Real policy trained via Natural Policy Gradient on MuJoCo simulation with randomized perturbations learns to walk in clutter and step over objects
Outdoor: gravel & branches
A Sim2Real policy trained via Natural Policy Gradient on MuJoCo simulation with randomized height field learns to walk outdoors over gravel and branches
Outdoor: slope & grass
A Sim2Real policy trained via Natural Policy Gradient on MuJoCo simulation with randomized height field learns to handle moderate slopes
Avoid Moving Obstacles
Avoid Moving Obstacles
Policy trained via Hierarchical Sim2Real learns to avoid a moving block and reach the target (marked by the controller on the floor)
Push to Moving Goal
Push to Moving Goal
Policy trained via Hierarchical Sim2Real learns to push block towards a moving target (marked by the controller in the hand)
Co-ordinate
Co-ordinate
Policy trained via Hierarchical Sim2Real learns to coordinate tow D'Kitties to push a heavy block towards a target (marked by two + signs on the floor)