Estimator-Coupled Reinforcement Learning for Robust Purely Tactile In-Hand Manipulation

This paper identifies and addresses the problems with naively combining (reinforcement) learning-based controllers and state estimators for robotic in-hand manipulation. Specifically, we tackle the challenging task of purely tactile, goal-conditioned, dextrous in-hand reorientation with the hand pointing downwards. Due to the limited sensing available, many control strategies that are feasible in simulation when having full knowledge of the object's state do not allow for accurate state estimation. Hence, separately training the controller and the estimator and combining the two at test time leads to poor performance. We solve this problem by coupling the control policy to the state estimator already during training in simulation. This approach leads to more robust state estimation and overall higher performance on the task while maintaining an interpretability advantage over end-to-end policy learning. With our GPU-accelerated implementation, learning from scratch takes a median training time of only 6.5 hours on a single, low-cost GPU. In simulation experiments with the DLR-Hand II and for four significantly different object shapes, we provide an in-depth analysis of the performance of our approach. We demonstrate the successful sim2real transfer by rotating the four objects to all 24 orientations in the $\pi/2$ discretization of SO(3), which has never been achieved for such a diverse set of shapes. Finally, our method can reorient a cube consecutively to nine goals (median), which was beyond the reach of previous methods in this challenging setting.

Self-Contained and Automatic Calibration of a Multi-Fingered Hand Using Only Pairwise Contact Measurements

A self-contained calibration procedure that can be performed automatically without additional external sensors or tools is a significant advantage, especially for complex robotic systems. Here, we show that the kinematics of a multi-fingered robotic hand can be precisely calibrated only by moving the tips of the fingers pairwise into contact. The only prerequisite for this is sensitive contact detection, e.g., by torque-sensing in the joints (as in our DLR-Hand II) or tactile skin. The measurement function for a given joint configuration is the distance between the modeled fingertip geometries, but the actual measurement is always zero. In an in-depth analysis, we prove that this contact-based calibration determines all quantities needed for manipulating objects with the hand, i.e., the difference vectors of the fingertips, and that it is as sensitive as a calibration using an external visual tracking system and markers. We describe the complete calibration scheme, including the selection of optimal sample joint configurations and search motions for the contacts despite the initial kinematic uncertainties. In a real-world calibration experiment for the torque-controlled four-fingered DLR-Hand II, the maximal error of 17.7mm can be reduced to only 3.7mm.

Dextrous Tactile In-Hand Manipulation Using a Modular Reinforcement Learning Architecture

Dextrous in-hand manipulation with a multi-fingered robotic hand is a challenging task, esp. when performed with the hand oriented upside down, demanding permanent force-closure, and when no external sensors are used. For the task of reorienting an object to a given goal orientation (vs. infinitely spinning it around an axis), the lack of external sensors is an additional fundamental challenge as the state of the object has to be estimated all the time, e.g., to detect when the goal is reached. In this paper, we show that the task of reorienting a cube to any of the 24 possible goal orientations in a ${\pi}$/2-raster using the torque-controlled DLR-Hand II is possible. The task is learned in simulation using a modular deep reinforcement learning architecture: the actual policy has only a small observation time window of 0.5s but gets the cube state as an explicit input which is estimated via a deep differentiable particle filter trained on data generated by running the policy. In simulation, we reach a success rate of 92% while applying significant domain randomization. Via zero-shot Sim2Real-transfer on the real robotic system, all 24 goal orientations can be reached with a high success rate.

Learning Purely Tactile In-Hand Manipulation with a Torque-Controlled Hand

We show that a purely tactile dextrous in-hand manipulation task with continuous regrasping, requiring permanent force closure, can be learned from scratch and executed robustly on a torque-controlled humanoid robotic hand. The task is rotating a cube without dropping it, but in contrast to OpenAI's seminal cube manipulation task, the palm faces downwards and no cameras but only the hand's position and torque sensing are used. Although the task seems simple, it combines for the first time all the challenges in execution as well as learning that are important for using in-hand manipulation in real-world applications. We efficiently train in a precisely modeled and identified rigid body simulation with off-policy deep reinforcement learning, significantly sped up by a domain adapted curriculum, leading to a moderate 600 CPU hours of training time. The resulting policy is robustly transferred to the real humanoid DLR Hand-II, e.g., reaching more than 46 full 2${\pi}$ rotations of the cube in a single run and allowing for disturbances like different cube sizes, hand orientation, or pulling a finger.