Learning to use chopsticks in diverse gripping styles

Abstract

Learning dexterous manipulation skills is a long-standing challenge in computer graphics and robotics, especially when the task involves complex and delicate interactions between the hands, tools and objects. In this paper, we focus on chopsticks-based object relocation tasks, which are common yet demanding. The key to successful chopsticks skills is steady gripping of the sticks that also supports delicate maneuvers. We automatically discover physically valid chopsticks holding poses by Bayesian Optimization (BO) and Deep Reinforcement Learning (DRL), which works for multiple gripping styles and hand morphologies without the need of example data. Given as input the discovered gripping poses and desired objects to be moved, we build physics-based hand controllers to accomplish relocation tasks in two stages. First, kinematic trajectories are synthesized for the chopsticks and hand in a motion planning stage. The key components of our motion planner include a grasping model to select suitable chopsticks configurations for grasping the object, and a trajectory optimization module to generate collision-free chopsticks trajectories. Then we train physics-based hand controllers through DRL again to track the desired kinematic trajectories produced by the motion planner. We demonstrate the capabilities of our framework by relocating objects of various shapes and sizes, in diverse gripping styles and holding positions for multiple hand morphologies. Our system achieves faster learning speed and better control robustness, when compared to vanilla systems that attempt to learn chopstick-based skills without a gripping pose optimization module and/or without a kinematic motion planner. Our code and models are available at this link. 1

Locomotion-Action-Manipulation: Synthesizing Human-Scene Interactions in Complex 3D Environments

“…Later work [15,70] extends this by synthesizing motions conditioned with variations of objects and contact points. Other approaches [47,54,55,66,67] focus on generating natural hand movements for manipulation, which is extended by including full body motions [54]. Physics-based character control to synthesize human object interactions has been also explored in [8,10,39,47,66].…”

“…Other approaches [47,54,55,66,67] focus on generating natural hand movements for manipulation, which is extended by including full body motions [54]. Physics-based character control to synthesize human object interactions has been also explored in [8,10,39,47,66]. Although these approaches cover a wide range of human object interactions, most of them solely focus on the relationship between human and the target object without long-term navigation in cluttered 3D scenes.…”

CAMS: CAnonicalized Manipulation Spaces for Category-Level Functional Hand-Object Manipulation Synthesis

“…Generating high-quality human grasps remains challenging due to the complex geometry and complicated skeletal constraints. Physics-based methods [7,23,28,33,39,40] were favored for in-hand manipulation synthesis since the generated motion was physically plausible. IBS [33] presented a novel representation of hand-object interaction and leveraged reinforcement learning (RL) methods with execution success, and geometric measure [11] rewards to generate successful grasping motion.…”

“…D-Grasp [7] developed its grasping policy based on the physical attributes of hand and object, including angles and velocities. Yang et al [39] concentrated on using chopsticks in diverse gripping styles, and it solved this rather difficult task by first optimizing physically valid gripping poses with predefined gripping styles and then utilizing the carefully designed hand-controlled policies to synthesize manipulation.…”

Bidirectional GaitNet: A Bidirectional Prediction Model of Human Gait and Anatomical Conditions

“…Recently, deep reinforcement learning (DRL) has successfully demonstrated its capabilities in solving high-dimensional, continuous control problems including human motion imitation [Bergamin et al 2019;Merel et al 2019;Peng et al 2018Peng et al , 2022Peng et al , 2021Won et al 2020Yu et al 2019], motion control in complex environments [Clegg et al 2018;Liu and Hodgins 2018;Winkler et al 2022;Won et al 2021;Yang et al 2022;] and non-human character control [Ishiwaka et al 2022;Lee et al 2022;Luo et al 2020;]. The control of musculoskeletal characters is no exception for these technological innovations; in particular, controllers based on DRL have been significantly improved in terms of robustness against external perturbation, computational efficiency at runtime, and the scope of reproducible motor skills.…”