One of the main challenges in peg-in-a-hole (PiH) insertion tasks is in handling the uncertainty in the location of the target hole. In order to address it, high-dimensional sensor inputs from sensor modalities such as vision, force/torque sensing, and proprioception can be combined to learn control policies that are robust to this uncertainty in the target pose. Whereas deep learning has shown success in recognizing objects and making decisions with high-dimensional inputs, the learning procedure might damage the robot when applying directly trial- and-error algorithms on the real system. At the same time, learning from Demonstration (LfD) methods have been shown to achieve compelling performance in real robotic systems by leveraging demonstration data provided by experts. In this paper, we investigate the merits of multiple sensor modalities such as vision, force/torque sensors, and proprioception when combined to learn a controller for real world assembly operation tasks using LfD techniques. The study is limited to PiH insertions; we plan to extend the study to more experiments in the future. Additionally, we propose a multi-step-ahead loss function to improve the performance of the behavioral cloning method. Experimental results on a real manipulator support our findings, and show the effectiveness of the proposed loss function.
Yaskawa India Pvt. Ltd.（安川印度私人有限公司）；Robert Bosch Center for Cyber-Physical Systems, Indian Institute of Science（印度科学学院罗伯特·博世信息物理系统中心）
Industrial robot manipulators are not able to match the precision and speed with which humans are able to execute contact rich tasks even to this day. Therefore, as a means overcome this gap, we demonstrate generative methods for imitating a peg-in-hole insertion task in a 6-DOF robot manipulator. In particular, generative adversarial imitation learning (GAIL) is used to successfully achieve this task with a 10 um, and a 6 um peg-hole clearance on the Yaskawa GP8 industrial robot. Experimental results show that the policy successfully learns within 20 episodes from a handful of human expert de…
The accurate modeling of real-world systems and physical interactions is a common challenge towards the resolution of robotics tasks. Machine learning approaches have demonstrated significant results in the modeling of complex systems (e.g., articulated robot structures, cable stretch, fluid dynamics), or to learn robotics tasks (e.g., grasping, reaching) from raw sensor measurements without explicit programming, using reinforcement learning. However, a common bottleneck in machine learning techniques resides in the availability of suitable data. While many vision-based datasets have been rele…
Learning generalizable insertion skills in a data-efficient manner has long been a challenge in the robot learning community. While the current state-of-the-art methods with reinforcement learning (RL) show promising performance in acquiring manipulation skills, the algorithms are data-hungry and hard to generalize. To overcome the issues, in this paper we present Prim-LAfD, a simple yet effective framework to learn and adapt primitive-based insertion skills from demonstrations. Prim-LAfD utilizes black-box function optimization to learn and adapt the primitive parameters leveraging prior expe…
Many industrial assembly tasks involve peg-in-hole like insertions with sub-millimeter tolerances which are challenging, even in highly calibrated robot cells. Visual servoing can be employed to increase the robustness towards uncertainties in the system, however, state of the art methods either rely on accurate 3D models for synthetic renderings or manual involvement in acquisition of training data. We present a novel self-supervised visual servoing method for high precision peg-in-hole insertion, which is fully automated and does not rely on synthetic data. We demonstrate its applicability f…
Even though the peg-hole insertion is one of the well-studied problems in robotics, it still remains a challenge for robots, especially when it comes to flexibility and the ability to generalize. Successful completion of the task requires combining several modalities to cope with the complexity of the real world. In our work, we focus on the visual aspect of the problem and employ the strategy of learning an insertion task in a simulator. We use Deep Reinforcement Learning to learn the policy end-to-end and then transfer the learned model to the real robot, without any additional fine-tuning. …
UC Berkeley（加州大学伯克利分校）；Karlsruhe Institute of Technology（卡尔斯鲁厄理工学院）；Autodesk（欧特克）；Autodesk Research（欧特克研究院）
Industrial insertion tasks are often performed repetitively with parts that are subject to tight tolerances and prone to breakage. Learning an industrial insertion policy in real is challenging as the collision between the parts and the environment can cause slippage or breakage of the part. In this paper, we present a safe self-supervised method to learn a visuo-tactile insertion policy that is robust to grasp pose variations. The method reduces human input and collisions between the part and the receptacle. The method divides the insertion task into two phases. In the first align phase, a ta…
SDU Robotics（南丹麦大学机器人中心）；University of Southern Denmark（南丹麦大学）
This paper demonstrates a visual servoing method which is robust towards uncertainties related to system calibration and grasping, while significantly reducing the peg-in-hole time compared to classical methods and recent attempts based on deep learning. The proposed visual servoing method is based on peg and hole point estimates from a deep neural network in a multi-cam setup, where the model is trained on purely synthetic data. Empirical results show that the learnt model generalizes to the real world, allowing for higher success rates and lower cycle times than existing approaches.
We propose a method that actively estimates contact location between a grasped rigid object and its environment and uses this as input to a peg-in-hole insertion policy. An estimation model and an active tactile feedback controller work collaboratively to estimate the external contacts accurately. The controller helps the estimation model get a better estimate by regulating a consistent contact mode. The better estimation makes it easier for the controller to regulate the contact. We then train an object-agnostic insertion policy that learns to use the series of contact estimates to guide the …
Mitsubishi Electric Research Laboratories (MERL)（三菱电机研究实验室）
We present the design of a learning-based compliance controller for assembly operations for industrial robots. We propose a solution within the general setting of learning from demonstration (LfD), where a nominal trajectory is provided through demonstration by an expert teacher. This can be used to learn a suitable representation of the skill that can be generalized to novel positions of one of the parts involved in the assembly, for example the hole in a peg-in-hole (PiH) insertion task. Under the expectation that this novel position might not be entirely accurately estimated by a vision or …
From Reach to Insert: Tactile-Augmented Precision Assembly under Sub-Millimeter Tolerances
State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所多模态人工智能系统国家重点实验室）；School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）
Carnegie Mellon University（卡内基梅隆大学）；Intrinsic（Intrinsic（谷歌旗下机器人公司））；Arizona State University（亚利桑那州立大学）
In this paper, we discuss a framework for teaching bimanual manipulation tasks by imitation. To this end, we present a system and algorithms for learning compliant and contact-rich robot behavior from human demonstrations. The presented system combines insights from admittance control and machine learning to extract control policies that can (a) recover from and adapt to a variety of disturbances in time and space, while also (b) effectively leveraging physical contact with the environment. We demonstrate the effectiveness of our approach using a real-world insertion task involving multiple si…
A Visual Reinforcement Learning-Based Separate Primitive Policy for Peg-in-Hole Tasks
State Key Laboratory of Robotics and Systems, Harbin Institute of Technology（机器人系统国家重点实验室，哈尔滨工业大学）；Ubtech Robotics（优必选科技）；Meituan Academy of Robotics（美团机器人研究院）；School of Mechanical Engineering, Shandong University（山东大学机械工程学院）
Complicated assembly processes can be described as a sequence of two main activities: grasping and insertion. While general grasping solutions are common in industry, insertion is still only applicable to small subsets of problems, mainly ones involving simple shapes in fixed locations and in which the variations are not taken into consideration. Recently, RL approaches with prior knowledge (e.g., LfD or residual policy) have been adopted. However, these approaches might be problematic in contact-rich tasks since interaction might endanger the robot and its equipment. In this paper, we tackled…
Spacetime Optimal-Transport Attention for Visuo-Haptic Imitation Learning of Contact-Rich Manipulation
Robotics Research Centre, School of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore（机器人研究中心，机械与航空航天工程学院，南洋理工大学，新加坡）；Wings Robotics（Wings机器人）
Robot manipulation requires a complex set of skills that need to be carefully combined and coordinated to solve a task. Yet, most ReinforcementLearning (RL) approaches in robotics study tasks which actually consist only of a single manipulation skill, such as grasping an object or inserting a pre-grasped object. As a result the skill ('how' to solve the task) but not the actual goal of a complete manipulation ('what' to solve) is specified. In contrast, we study a complex manipulation goal that requires an agent to learn and combine diverse manipulation skills. We propose a challenging, highly…
Osaka University（大阪大学）；National Institute of Advanced Industrial Science and Technology (AIST)（日本产业技术综合研究所）
Industrial robot manipulators are playing a more significant role in modern manufacturing industries. Though peg-in-hole assembly is a common industrial task which has been extensively researched, safely solving complex high precision assembly in an unstructured environment remains an open problem. Reinforcement Learning (RL) methods have been proven successful in solving manipulation tasks autonomously. However, RL is still not widely adopted on real robotic systems because working with real hardware entails additional challenges, especially when using position-controlled manipulators. The ma…
We present a learning-based force-torque dynamics to achieve model-based control for contact-rich peg-in-hole task using force-only inputs. Learning the force-torque dynamics is challenging because of the ambiguity of the low-dimensional 6-d force signal and the requirement of excessive training data. To tackle these problems, we propose a multi-pose force-torque state representation, based on which a dynamics model is learned with the data generated in a sample-efficient offline fashion. In addition, by training the dynamics model with peg-and-holes of various shapes, scales, and elasticities…
Osaka University（大阪大学）；National Inst. of AIST（国立先进工业科学技术研究所）
This paper uses robots to assemble pegs into holes on surfaces with different colors and textures. It especially targets at the problem of peg-in-hole assembly with initial position uncertainty. Two in-hand cameras and a force-torque sensor are used to account for the position uncertainty. A program sequence comprising learning-based visual servoing, spiral search, and impedance control is implemented to perform the peg-in-hole task with feedback from the above sensors. Contributions are mainly made in the learning-based visual servoing of the sequence, where a deep neural network is trained w…
Rainbow Deep Q-Learning with Kinematics-Aware Design for Cooperative Delta and 3-RRS Parallel Robot Insertion
Robotics Research Center of Yuyao（余姚机器人研究中心）；Robotics Institute of Zhejiang University（浙江大学机器人院）；Yuyao Technology Innovation Center（余姚技术创新中心）；Department of Electrical and Computer Engineering, University of New Brunswick（新 Brunswick大学电气与计算机工程系）
One Demonstration Is Enough for Real-World Robotic Reinforcement Learning
National Key Laboratory of Cognition and Decision Intelligence for Complex Systems, Institution of Automation, Chinese Academy of Sciences（中国科学院自动化研究所复杂系统认知与决策智能国家重点实验室）；Beijing Academy of Artificial Intelligence（北京智源人工智能研究院）；PKU-PsiBot Joint Lab（北大-鹏城实验室联合实验室）；School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）；Institute for Artificial Intelligence, Peking University（北京大学人工智能研究院）
In the peg insertion task, human pays attention to the seam between the peg and the hole and tries to fill it continuously with visual feedback. By imitating the human behavior, we design architectures with position and orientation estimators based on the seam representation for pose alignment, which proves to be general to the unseen peg geometries. By putting the estimators into the closed-loop control with reinforcement learning, we further achieve a higher or comparable success rate, efficiency, and robustness compared with the baseline methods. The policy is trained totally in simulation …
VTLA: Vision-Tactile-Language-Action Model with Preference Learning for Insertion Manipulation
State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences（多模态人工智能系统国家重点实验室，自动化研究所，中国科学院）；Samsung R&D Institute China–Beijing（三星中国北京研发中心）；Beijing Academy of Artificial Intelligence（北京人工智能研究院）
Manipulation tasks often require a robot to adjust its sensorimotor skills based on the state it finds itself in. Taking peg-in-hole as an example: once the peg is aligned with the hole, the robot should push the peg downwards. While high level execution frameworks such as state machines and behavior trees are commonly used to formalize such decision-making problems, these frameworks require a mechanism to detect the high-level symbolic state. Handcrafting heuristics to identify symbolic states can be brittle, and using data-driven methods can produce noisy predictions, particularly when worki…
Insertion is a challenging haptic and visual control problem with significant practical value for manufacturing. Existing approaches in the model-based robotics community can be highly effective when task geometry is known, but are complex and cumbersome to implement, and must be tailored to each individual problem by a qualified engineer. Within the learning community there is a long history of insertion research, but existing approaches are typically either too sample-inefficient to run on real robots, or assume access to high-level object features, e.g. socket pose. In this paper we show th…
College of Computer and Information Science, Northeastern University（东北大学计算机与信息科学学院）
In applications of deep reinforcement learning to robotics, it is often the case that we want to learn pose invariant policies: policies that are invariant to changes in the position and orientation of objects in the world. For example, consider a peg-in-hole insertion task. If the agent learns to insert a peg into one hole, we would like that policy to generalize to holes presented in different poses. Unfortunately, this is a challenge using conventional methods. This paper proposes a novel state and action abstraction that is invariant to pose shifts called \textit{deictic image maps} that c…
InsertAnything: Generalizable Contact-Rich Precision Insertion from Simulation to Reality
University of Chinese Academy of Sciences（中国科学院大学）；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；PaXini AI Technology (Beijing) Co., Ltd.（北京他山科技有限公司）
The development of a generic and effective force model for semi-automatic or manual virtual assembly with haptic support is not a trivial task, especially when the assembly constraints involve complex features of arbitrary shape. The primary challenge lies in a proper formulation of the guidance forces and torques that effectively assist the user in the exploration of the virtual environment (VE), from repulsing collisions to attracting proper contact. The secondary difficulty is that of efficient implementation that maintains the standard 1 kHz haptic refresh rate. We propose a purely geometr…
Context-Based Meta Reinforcement Learning for Robust and Adaptable Peg-in-Hole Assembly Tasks
Humanoid Robots Lab and Center for Robotics, University of Bonn, Germany（波恩大学人形机器人实验室和机器人中心）；Lamarr Institute for Machine Learning and Artificial Intelligence, Bonn, Germany（波恩大学拉马尔人工智能与机器学习研究所）；Cyber Physical Systems Lab, Egypt Japan University of Science and Technology, Alexandria, Egypt（埃及日本科学技术大学网络物理系统实验室）；Faculty of Engineering, Alexandria University, Alexandria, Egypt（亚历山大大学工程学院）；Mechatronics Department, Ain Shams University, Cairo, Egypt（亚历山大大学机电系）
In many bi-manual robotic tasks, like peg-in-a-hole assembly, the success of the task execution depends on the error in achieving the desired relative pose between the peg and the hole in a pre-insertion configuration. Random actuation errors in the joint space usually prevent the two arms from reaching their desired task space poses, which in turn results in a random error in relative pose between the two hands. This random error varies from trial to trial, and thus depending on the tolerance between the peg and the hole, the outcome of the assembly task may be random (sometimes the task exec…
University of Colorado, Boulder（科罗拉多大学博尔德分校）；Robotic Materials Inc.（机器人材料公司）
We present algorithms and results for a robotic manipulation system that was designed to be easily programmable and adaptable to various tasks common to industrial setting, which is inspired by the Industrial Assembly Challenge at the 2018 World Robotics Summit in Tokyo. This challenge included assembly of standard, commercially available industrial parts into 2D and 3D assemblies. We demonstrate three tasks that can be classified into "peg-in-hole" and "hole-on-peg" tasks and identify two canonical algorithms: spiral-based search and tilting insertion. Both algorithms use hand-coded threshold…
Highly constrained manipulation tasks continue to be challenging for autonomous robots as they require high levels of precision, typically less than 1mm, which is often incompatible with what can be achieved by traditional perception systems. This paper demonstrates that the combination of state-of-the-art object tracking with passively adaptive mechanical hardware can be leveraged to complete precision manipulation tasks with tight, industrially-relevant tolerances (0.25mm). The proposed control method closes the loop through vision by tracking the relative 6D pose of objects in the relevant …
