Robotic Manipulation
Robotic manipulation research focuses on enabling robots to dexterously interact with their environment, achieving complex tasks through precise movements and object manipulation. Current efforts concentrate on improving the robustness and generalization of manipulation policies, often leveraging vision-language models, transformer architectures, and reinforcement learning techniques to enable robots to understand and respond to diverse instructions and environments. This field is crucial for advancing automation in various sectors, from manufacturing and logistics to healthcare and agriculture, by creating more adaptable and reliable robotic systems capable of handling a wider range of tasks. Furthermore, significant attention is being paid to developing more efficient data collection methods and improving the safety and reliability of these systems.
Papers
Graph Neural Networks for Relational Inductive Bias in Vision-based Deep Reinforcement Learning of Robot Control
Marco Oliva, Soubarna Banik, Josip Josifovski, Alois Knoll
6-DoF Pose Estimation of Household Objects for Robotic Manipulation: An Accessible Dataset and Benchmark
Stephen Tyree, Jonathan Tremblay, Thang To, Jia Cheng, Terry Mosier, Jeffrey Smith, Stan Birchfield
AW-Opt: Learning Robotic Skills with Imitation and Reinforcement at Scale
Yao Lu, Karol Hausman, Yevgen Chebotar, Mengyuan Yan, Eric Jang, Alexander Herzog, Ted Xiao, Alex Irpan, Mohi Khansari, Dmitry Kalashnikov, Sergey Levine
Robot control for simultaneous impact tasks via Quadratic Programming-based reference spreading
Jari J. van Steen, Nathan van de Wouw, Alessandro Saccon