ETD

• grasping
• machine learning
• object decomposition
• planning
• robotics
• soft robotics

The goal of this thesis is have a robot that learn how to grasp an unknown object with an underactuated end-effector. We chose to implement regression via deep convolutional neural network in a supervised learning scenario. We present a 3D and 2D deep convolutional neural network with as input, respectively a voxel and images takes from vary point of view of the object.

We create a database that match the image with one feasible pose. The pose is make using Minimum Volume Bounding Box, minimizing the volume of the boxes which fit partial point clouds. In these way we can focus on outermost boxes and we can choose a desired pose that grasping the object in a successful manner.

The simulation is made using Klamp't simulator, and the neural network is implement using Theano library.

We compare the results of our neural network trying different loss function and two structure of network: one with regularization and the other without it.