Rico Jonschkowski

Learn More
State representations critically affect the effectiveness of learning in robots. In this paper, we propose a roboticsspecific approach to learning such state representations. Robots accomplish tasks by interacting with the physical world. Physics in turn imposes structure on both the changes in the world and on the way robots can effect these changes. Using(More)
We describe the winning entry to the Amazon Picking Challenge. From the experience of building this system and competing in the Amazon Picking Challenge, we derive several conclusions: 1) We suggest to characterize robotic system building along four key aspects, each of them spanning a spectrum of solutions—modularity vs. integration, generality vs.(More)
We describe the winning entry to the Amazon Picking Challenge. From the experience of building this system and competing in the Amazon Picking Challenge, we derive several conclusions: 1) We suggest to characterize robotic systems building along four key aspects, each of them spanning a spectrum of solutions—modularity vs. integration, generality vs.(More)
The success of reinforcement learning in robotic tasks is highly dependent on the state representation – a mapping from high dimensional sensory observations of the robot to states that can be used for reinforcement learning. Even though many methods have been proposed to learn state representations, it remains an important open problem. Identifying the(More)
We present a method for multi-class segmentation from RGB-D data in a realistic warehouse picking setting. The method computes pixel-wise probabilities and combines them to find a coherent object segmentation. It reliably segments objects in cluttered scenarios, even when objects are translucent, reflective, highly deformable, have fuzzy surfaces, or(More)
Problem-specific robotic algorithms and generic machine learning approaches to robotics have complementary strengths and weaknesses, trading-off data-efficiency and generality. To find the right balance between these, we propose to use robotics-specific information encoded in robotic algorithms together with the ability to learn task-specific information(More)
We propose position-velocity encoders (PVEs) which learn—without supervision—to encode images to positions and velocities of task-relevant objects. PVEs encode a single image into a low-dimensional position state and compute the velocity state from finite differences in position. In contrast to autoencoders, position-velocity encoders are not trained by(More)