Lukás Hermann
YOU?
Author Swipe
View article: RobotIO: A Python Library for Robot Manipulation Experiments
RobotIO: A Python Library for Robot Manipulation Experiments Open
Setting up robot environments to quickly test newly developed algorithms is still a difficult and time consuming process. This presents a significant hurdle to researchers interested in performing real-world robotic experiments. RobotIO is…
View article: What Matters in Language Conditioned Robotic Imitation Learning over Unstructured Data
What Matters in Language Conditioned Robotic Imitation Learning over Unstructured Data Open
A long-standing goal in robotics is to build robots that can perform a wide range of daily tasks from perceptions obtained with their onboard sensors and specified only via natural language. While recently substantial advances have been ac…
View article: Affordance Learning from Play for Sample-Efficient Policy Learning
Affordance Learning from Play for Sample-Efficient Policy Learning Open
Robots operating in human-centered environments should have the ability to understand how objects function: what can be done with each object, where this interaction may occur, and how the object is used to achieve a goal. To this end, we …
View article: CALVIN: A Benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation Tasks
CALVIN: A Benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation Tasks Open
General-purpose robots coexisting with humans in their environment must learn to relate human language to their perceptions and actions to be useful in a range of daily tasks. Moreover, they need to acquire a diverse repertoire of general-…
View article: Pre-training of Deep RL Agents for Improved Learning under Domain Randomization
Pre-training of Deep RL Agents for Improved Learning under Domain Randomization Open
Visual domain randomization in simulated environments is a widely used method to transfer policies trained in simulation to real robots. However, domain randomization and augmentation hamper the training of a policy. As reinforcement learn…
View article: PIXLISE-C: Exploring The Data Analysis Needs of NASA Scientists for Mineral Identification
PIXLISE-C: Exploring The Data Analysis Needs of NASA Scientists for Mineral Identification Open
NASA JPL scientists working on the micro x-ray fluorescence (microXRF) spectroscopy data collected from Mars surface perform data analysis to look for signs of past microbial life on Mars. Their data analysis workflow mainly involves ident…
View article: Hindsight for Foresight: Unsupervised Structured Dynamics Models from Physical Interaction
Hindsight for Foresight: Unsupervised Structured Dynamics Models from Physical Interaction Open
A key challenge for an agent learning to interact with the world is to reason about physical properties of objects and to foresee their dynamics under the effect of applied forces. In order to scale learning through interaction to many obj…
View article: FlowControl: Optical Flow Based Visual Servoing
FlowControl: Optical Flow Based Visual Servoing Open
One-shot imitation is the vision of robot programming from a single demonstration, rather than by tedious construction of computer code. We present a practical method for realizing one-shot imitation for manipulation tasks, exploiting mode…
View article: Hindsight for Foresight: Unsupervised Structured Dynamics Models from\n Physical Interaction
Hindsight for Foresight: Unsupervised Structured Dynamics Models from\n Physical Interaction Open
A key challenge for an agent learning to interact with the world is to reason\nabout physical properties of objects and to foresee their dynamics under the\neffect of applied forces. In order to scale learning through interaction to\nmany …
View article: Adaptive Curriculum Generation from Demonstrations for Sim-to-Real Visuomotor Control
Adaptive Curriculum Generation from Demonstrations for Sim-to-Real Visuomotor Control Open
We propose Adaptive Curriculum Generation from Demonstrations (ACGD) for reinforcement learning in the presence of sparse rewards. Rather than designing shaped reward functions, ACGD adaptively sets the appropriate task difficulty for the …
View article: Adaptive Curriculum Generation from Demonstrations for Sim-to-Real\n Visuomotor Control
Adaptive Curriculum Generation from Demonstrations for Sim-to-Real\n Visuomotor Control Open
We propose Adaptive Curriculum Generation from Demonstrations (ACGD) for\nreinforcement learning in the presence of sparse rewards. Rather than designing\nshaped reward functions, ACGD adaptively sets the appropriate task difficulty\nfor t…