Learning Visuotactile Estimation and Control for Non-prehensile Manipulation under Occlusions

Juan Del Aguila Ferrandis, Joao Moura, Sethu Vijayakumar

Keywords: State Estimation, Reinforcement Learning with Tactile Sensing, Non-prehensile Manipulation

TL;DR: We propose a method to learn visuotactile state estimators and uncertainty-aware control policies for non-prehensile manipulation under occlusions and extensively validate it in simulation and robotic hardware.

Abstract: Manipulation without grasping, known as non-prehensile manipulation, is essential for dexterous robots in contact-rich environments, but presents many challenges relating with underactuation, hybrid-dynamics, and frictional uncertainty. Additionally, object occlusions in a scenario of contact uncertainty and where the motion of the object evolves independently from the robot becomes a critical problem, which previous literature fails to address. We present a method for learning visuotactile state estimators and uncertainty-aware control policies for non-prehensile manipulation under occlusions, by leveraging diverse interaction data from privileged policies trained in simulation. We formulate the estimator within a Bayesian deep learning framework, to model its uncertainty, and then train uncertainty-aware control policies by incorporating the pre-learned estimator into the reinforcement learning (RL) loop, both of which lead to significantly improved estimator and policy performance. Therefore, unlike prior non-prehensile research that relies on complex external perception set-ups, our method successfully handles occlusions after sim-to-real transfer to robotic hardware with a simple onboard camera.

Video: https://youtu.be/hW-C8i_HWgs

Student Paper: yes

Submission Number: 402