The field of robotic manipulation is rapidly advancing, with a focus on developing more sophisticated tactile sensing and manipulation capabilities. Recent research has explored the use of tactile sensing to improve robotic grasping and manipulation, including the development of new tactile sensors and algorithms for processing tactile data. Another area of focus has been on improving the robustness and adaptability of robotic manipulation systems, including the use of reinforcement learning and simulation-based training to enable robots to learn from experience and adapt to new situations. Notable papers in this area include the development of a novel robotic hand with adaptive grasping capabilities, a method for learning to throw-flip objects with a robot, and a framework for active tactile exploration for rigid body pose and shape estimation.

Noteworthy papers include: Cross-Sensor Touch Generation, which proposes two approaches to cross-sensor image generation, enabling the use of sensor-specific models across multiple sensors. Learning to Throw-Flip, which presents a method enabling a robot to accurately throw-flip objects to a desired landing pose. Refinery, which introduces a framework that bridges the performance gap in simulation-based learning for robotic assembly, robustifying policy performance across initial conditions.