Soft robots are gaining popularity thanks to their intrinsic safety to contacts and adaptability. However, the potentially infinite number of Degrees of Freedom makes their modeling a daunting task, and in many cases only an approximated description is available. This challenge makes reinforcement learning (RL) based approaches inefficient when deployed on a realistic scenario, due to the large domain gap between models and the real platform. In this work, we demonstrate, for the first time, how Domain Randomization (DR) can solve this problem by enhancing RL policies for soft robots with: i) robustness w.r.t. unknown dynamics parameters; ii) reduced training times by exploiting drastically simpler dynamic models for learning; iii) better environment exploration, which can lead to exploitation of environmental constraints for optimal performance. Moreover, we introduce a novel algorithmic extension to previous adaptive domain randomization methods for the automatic inference of dynamics parameters for deformable objects. We provide an extensive evaluation in simulation on four different tasks and two soft robot designs, opening interesting perspectives for future research on Reinforcement Learning for closed-loop soft robot control.
University of Cambridge（剑桥大学）；Helmut-Schmidt-University Hamburg（汉堡赫尔穆特·施密特大学）
Reinforcement learning frameworks have introduced abstractions to implement and execute algorithms at scale. They assume standardized simulator interfaces but are not concerned with identifying suitable task representations. We present Wield, a first-of-its kind system to facilitate task design for practical reinforcement learning. Through software primitives, Wield enables practitioners to decouple system-interface and deployment-specific configuration from state and action design. To guide experimentation, Wield further introduces a novel task design protocol and classification scheme centre…
Autonomous grasping remains challenging as unlike humans, robots do not possess a sophisticated sensing nor delicate interaction capability with the real environment. Among other efforts that tried to close the gap between them, anthropomorphic robotic hands is the most prominent direction. However, exactly following human hand design might be unnecessary as it will significantly increase the mechanical complexity and hence make it less economically feasible. Recently, soft robotic hands, a new trend has emerged, aiming to make the design adequately complex and affordable while requiring much …
Panasonic Beta（松下Beta）；Totemic Inc.（Totemic公司）；University of California Berkeley（加州大学伯克利分校）
In order to practically implement the door opening task, a policy ought to be robust to a wide distribution of door types and environment settings. Reinforcement Learning (RL) with Domain Randomization (DR) is a promising technique to enforce policy generalization, however, there are only a few accessible training environments that are inherently designed to train agents in domain randomized environments. We introduce DoorGym, an open-source door opening simulation framework designed to utilize domain randomization to train a stable policy. We intend for our environment to lie at the intersect…
University of California, San Diego（加利福尼亚大学圣迭戈分校）；Technical University of Denmark（丹麦技术大学）
Extensive efforts have been made to improve the generalization ability of Reinforcement Learning (RL) methods via domain randomization and data augmentation. However, as more factors of variation are introduced during training, optimization becomes increasingly challenging, and empirically may result in lower sample efficiency and unstable training. Instead of learning policies directly from augmented data, we propose SOft Data Augmentation (SODA), a method that decouples augmentation from policy learning. Specifically, SODA imposes a soft constraint on the encoder that aims to maximize the mu…
Modern navigation algorithms based on deep reinforcement learning (RL) show promising efficiency and robustness. However, most deep RL algorithms operate in a risk-neutral manner, making no special attempt to shield users from relatively rare but serious outcomes, even if such shielding might cause little loss of performance. Furthermore, such algorithms typically make no provisions to ensure safety in the presence of inaccuracies in the models on which they were trained, beyond adding a cost-of-collision and some domain randomization while training, in spite of the formidable complexity of th…
Istituto Italiano di Tecnologia (IIT)（意大利理工学院）；Politecnico di Bari（巴里理工大学）；University of Southern Denmark（南丹麦大学）
Soft robotics technology can aid in achieving United Nations Sustainable Development Goals (SDGs) and the Paris Climate Agreement through development of autonomous, environmentally responsible machines powered by renewable energy. By utilizing soft robotics, we can mitigate the detrimental effects of climate change on human society and the natural world through fostering adaptation, restoration, and remediation. Moreover, the implementation of soft robotics can lead to groundbreaking discoveries in material science, biology, control systems, energy efficiency, and sustainable manufacturing pro…
For several years, high development and production costs of humanoid robots restricted researchers interested in working in the field. To overcome this problem, several research groups have opted to work with simulated or smaller robots, whose acquisition costs are significantly lower. However, due to scale differences and imperfect simulation replicability, results may not be directly reproducible on real, adult-sized robots. In this paper, we present the NimbRo-OP2X, a capable and affordable adult-sized humanoid platform aiming to significantly lower the entry barrier for humanoid robot rese…
Team Starkit；Moscow Institute of Physics and Technology（莫斯科物理技术学院）
In this paper, we present a robot model and code base for affordable education in the field of humanoid robotics. We give an overview of the software and hardware of a robot that won several competitions with the team RoboKit in 2019-2021, provide analysis of the contemporary market of education in robotics, and highlight the reasoning beyond certain design solutions.
The use of standard platforms in the field of humanoid robotics can accelerate research, and lower the entry barrier for new research groups. While many affordable humanoid standard platforms exist in the lower size ranges of up to 60cm, beyond this the few available standard platforms quickly become significantly more expensive, and difficult to operate and maintain. In this paper, the igus Humanoid Open Platform is presented---a new, affordable, versatile and easily customisable standard platform for humanoid robots in the child-sized range. At 90cm, the robot is large enough to interact wit…
Rutgers University（罗格斯大学）；Massachusetts Institute of Technology（麻省理工学院）
microMVP is an affordable, portable, and open source micro-scale mobile robot platform designed for robotics research and education. As a complete and unique multi-vehicle platform enabled by 3D printing and the maker culture, microMVP can be easily reproduced and requires little maintenance: a set of six micro vehicles, each measuring $8\times 5\times 6$ cubic centimeters and weighing under $100$ grams, and the accompanying tracking platform can be fully assembled in under two hours, all from readily available components. In this paper, we describe microMVP's hardware and software architectur…
TAME-Nav: Temporal Asymmetric Mixture-of-Experts Distillation for Agile and Versatile Robot Navigation
Shanghai Jiao Tong University（上海交通大学）；Eastern Institute of Technology（东方理工大学）；The Hong Kong Polytechnic University（香港理工大学）；University of Science and Technology of China（中国科学技术大学）
