Jonatan Sjølund Dyrstad
Master of Science
Jonatan Sjølund Dyrstad
Master of Science
Publications and responsibilities
Teaching a robot to grasp real fish by imitation learning from a human supervisor in virtual reality
We teach a real robot to grasp real fish, by training a virtual robot exclusively in virtual reality. Our approach implements robot imitation learning from a human supervisor in virtual reality. A deep 3D convolutional neural network computes grasps from a 3D occupancy grid obtained from depth...
Bin Picking of Reflective Steel Parts Using a Dual-Resolution Convolutional Neural Network Trained in a Simulated Environment
We consider the case of robotic bin picking of reflective steel parts, using a structured light 3D camera as a depth imaging device. In this paper, we present a new method for bin picking, based on a dual-resolution convolutional neural network trained entirely in a simulated environment. The dual...
Bin Picking of Reflective Steel Parts using a Dual-Resolution Convolutional Neural Network Trained in a Simulated Environment
Grasping virtual fish: A step towards deep learning from demonstration in virtual reality
We present an approach to robotic deep learning from demonstration in virtual reality, which combines a deep 3D convolutional neural network, for grasp detection from 3D point clouds, with domain randomization to generate a large training data set. The use of virtual reality (VR) enables robot...
Learning robots in the seafood industry
Simplifying automation in the food industry using deep learning and virtual reality
Grasping Virtual Fish - A Step Towards Robotic Deep Learning from Demonstration in Virtual Reality
Identifikasjon av lakseindivider — Biometri fase 1 (SalmID)
I lakseindustrien er fleksibilitet svært viktig for å kunne levere et mangfold av produkter til konsumentene, på en effektiv og lønnsom måte. På samme tid ønskes det å utnytte råvaren (fisken) på den beste og mest optimale måten for mest mulig høykvalitets produkter som kan gå til forbrukeren...