☆ 4.7 Review

Obstacles and opportunities for learning from demonstration in practical industrial assembly: A systematic literature review

ROBOTICS AND COMPUTER-INTEGRATED MANUFACTURING (2024)

相关参考文献

注意：仅列出部分参考文献，下载原文获取全部文献信息。

Article Automation & Control Systems

An Efficient Robot Precision Assembly Skill Learning Framework Based on Several Demonstrations

Yanqin Ma et al.

Summary: This paper proposes an efficient robot assembly skill learning framework based on only a few demonstrations. The framework consists of a pre-training phase and a self-learning phase, and utilizes novel data augmentation and hybrid exploration strategies to improve learning efficiency and skill performance.

IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING (2023)

添加到收藏夹

Article Automation & Control Systems

Online task segmentation by merging symbolic and data-driven skill recognition during kinesthetic teaching?

Thomas Eiband et al.

Summary: Programming by Demonstration (PbD) is used to transfer a task from a human teacher to a robot, and understanding the underlying structure of the demonstrated task is crucial. This paper proposes a framework that recognizes skills online during a kinesthetic demonstration, independent of visual perception. The framework achieves a 96% accuracy in skill recognition and highlights the benefits of the generated task representation compared to a baseline representation.

ROBOTICS AND AUTONOMOUS SYSTEMS (2023)

添加到收藏夹

Article Automation & Control Systems

Robotic assembly strategy via reinforcement learning based on force and visual information

Kuk-Hyun Ahn et al.

Summary: Robotic assembly strategies are typically limited to structured environments with fixed positions, but this study proposes a generalized assembly strategy based on force and visual information. Combining neural network trajectory generators and deep reinforcement learning, the robot can successfully accomplish assembly tasks regardless of shape or initial error.

ROBOTICS AND AUTONOMOUS SYSTEMS (2023)

添加到收藏夹

Article Automation & Control Systems

A robot learning from demonstration framework for skillful small parts assembly

Haopeng Hu et al.

Summary: The increasing demand for flexibility in production and smaller batch sizes has led the manufacturing industry to adopt robotic solutions that offer quick setup and reprogramming capabilities. The learning from demonstration (LfD) paradigm has gained attention as a way to facilitate the integration of robots in assembly lines. In this study, a robot LfD framework is developed for skillful small parts assembly applications, considering position, orientation, and wrench demonstration data. Two cascaded assembly policies are learned from separate assembly demonstrations to prevent potential under-fitting issues in industrial small parts assembly applications. The effectiveness of the proposed LfD framework is validated through a printed circuit board assembly experiment with a torque-controlled robot.

INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY (2022)

添加到收藏夹

Article Computer Science, Artificial Intelligence

A Framework of Improving Human Demonstration Efficiency for Goal-Directed Robot Skill Learning

Hongmin Wu et al.

Summary: Robot learning from humans allows robots to adapt to stochastic and dynamic environments through demonstrations provided by non-technical users. A goal-directed robot skill learning framework named GP(m)-MOGP is proposed to reduce the effort of non-technical users and improve the efficiency and success rate of robot learning. The framework decides when and where to add new demonstrations, determines the usefulness of demonstrations, and uses a multioutput Gaussian process for skill learning.

IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS (2022)

添加到收藏夹

Article Robotics

Residual Learning From Demonstration: Adapting DMPs for Contact-Rich Manipulation

Todor Davchev et al.

Summary: This paper explores how robots can learn manipulation skills involving contact and friction. The authors propose a combination of Dynamic Movement Primitives (DMP) and residual learning, which significantly improves the overall performance of DMPs. The evaluation results demonstrate the effectiveness of the proposed framework.

IEEE ROBOTICS AND AUTOMATION LETTERS (2022)

添加到收藏夹

Article Robotics

Ergodic Exploration Using Tensor Train: Applications in Insertion Tasks

Suhan Shetty et al.

Summary: Ergodic control in robotics extends the tracking principle by specifying a probability distribution over an area to cover, rather than tracking a trajectory. The use of tensor trains effectively addresses challenges in spatial distribution decomposition in control problems, and is suitable for online implementation in robotic systems.

IEEE TRANSACTIONS ON ROBOTICS (2022)

添加到收藏夹

Article Robotics

Deep latent-space sequential skill chaining from incomplete demonstrations

Minjae Kang et al.

Summary: Imitation learning is a method to train agents using demonstrations from experts without external rewards. It is challenging to perform complex tasks with independently trained low-level skills, which can be addressed by connecting skills and designing reward functions, as well as using binary classifiers.

INTELLIGENT SERVICE ROBOTICS (2022)

添加到收藏夹

Article Robotics

Learning Accurate and Stable Point-to-Point Motions: A Dynamic System Approach

Yu Zhang et al.

Summary: This study proposes a dynamic system approach for learning point-to-point motions while maintaining system stability. The approach is based on a neural network learning from demonstration method, which provides better reproduction performance and generalization ability. Experimental results show that the approach achieves higher accuracy, can handle high dimensional data, and is robust to variations in starting points.

IEEE ROBOTICS AND AUTOMATION LETTERS (2022)

添加到收藏夹

Article Automation & Control Systems

A Reconfigurable Multirobot Cooperation Workcell for Personalized Manufacturing

Bidan Huang et al.

Summary: This article introduces a robotic system for manufacturing personalized medical stent graft implants based on a modular design, which utilizes real-time 3D vision, multirobot collaboration, and personalization to guide robots to adapt to different implant geometry. The system can efficiently generate collision-free paths and is generalizable to different stent graft designs.

IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING (2022)

添加到收藏夹

Review Engineering, Multidisciplinary

Intelligent learning model-based skill learning and strategy optimization in robot grinding and polishing

Chen Chen et al.

Summary: With the rapid advancement of manufacturing in China, robot machining technology has become a popular research subject. Skill learning and strategy optimization have become the focus of research, and intelligent learning models have been widely used. This paper summarizes the state-of-the-art in skill learning and strategy optimization research and discusses future development trends.

SCIENCE CHINA-TECHNOLOGICAL SCIENCES (2022)

添加到收藏夹

Article Robotics

Learning to Pick at Non-Zero-Velocity From Interactive Demonstrations

Anna Meszaros et al.

Summary: This study investigates how to learn the intricate task of continuous pick & place motion from human demonstrations and corrections. The framework allows non-expert users to modify the dynamics of their initial demonstration through teleoperated corrective feedback, enabling them to teach motions beyond their physical capabilities. The framework learns the desired movement dynamics based on current Cartesian position with Gaussian Processes, allowing for online interactive corrections and active disturbance rejection.

IEEE ROBOTICS AND AUTOMATION LETTERS (2022)

添加到收藏夹

Review Robotics

Learning from Demonstrations in Human-Robot Collaborative Scenarios: A Survey

Arturo Daniel Sosa-Ceron et al.

Summary: Human-Robot Collaboration is an interdisciplinary research area with significance in smart manufacturing. By combining the physical capabilities of robots with the cognitive abilities of humans, tasks can be designed with efficiency and adaptability. Learning from Demonstration (LfD) is a promising programming technique that simplifies tasks and involves the shop floor operator. However, human intervention and participation have been poorly explored in the research.

ROBOTICS (2022)

添加到收藏夹

Article Automation & Control Systems

Learning to Assemble Noncylindrical Parts Using Trajectory Learning and Force Tracking

Jianhua Su et al.

Summary: This article introduces a method to teach a robot assembly skills using learning from demonstration, training both trajectory and insertion force simultaneously. By encoding human demonstrations data via motion primitives and using an adaptive impedance controller to track prescribed force, the proposed approach combines adaptive impedance control techniques with learning from demonstration, proving efficient in experiments on several typical noncylindrical parts.

IEEE-ASME TRANSACTIONS ON MECHATRONICS (2022)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Robotic Manipulation Skill Acquisition Via Demonstration Policy Learning

Dong Liu et al.

Summary: Current robots are constrained by environment and task variations, but teaching robots by demonstration is an effective approach. A learning-by-imitation method is proposed to learn robotic manipulation skills from interaction data, improving adaptability with fewer training inputs.

IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS (2022)

添加到收藏夹

Article Robotics

Perturbation-Based Stiffness Inference in Variable Impedance Control

Edoardo Caldarelli et al.

Summary: This work presents a force-based algorithm for inferring time-varying robot stiffness profiles through human-robot interaction in demonstration learning. Experimental results show that the algorithm can be combined with different types of force sensors, providing suitable processing algorithms to accurately infer stiffness profiles.

IEEE ROBOTICS AND AUTOMATION LETTERS (2022)

添加到收藏夹

Proceedings Paper Automation & Control Systems

Experimental Evaluation Of Intuitive Programming Of Robot Interaction Behaviour During Kinesthetic Teaching Using sEMG And Cutaneous Feedback

Roberto Meattini et al.

Summary: Modern applications can benefit from collaborative robots with advanced features, but smooth interactions cannot be intuitively programmed. This study proposes a robot programming framework that allows simultaneous teaching of trajectories and interaction behavior through a wearable interface. The use of surface electromyography and vibrotactile stimulation enables offline programming of robot trajectory and interaction behavior, providing operators with the ability to plan interactions with the environment and humans.

IFAC PAPERSONLINE (2022)

添加到收藏夹

Proceedings Paper Automation & Control Systems

Experimental Evaluation Of Intuitive Programming Of Robot Interaction Behaviour During Kinesthetic Teaching Using sEMG And Cutaneous Feedback

Roberto Meattini et al.

Summary: This study proposes a framework for robot trajectory and interaction behavior programming through kinesthetic teaching and a wearable interface. Experimental results show that the operator successfully utilized the provided wearable interface, and offline intuitive programming can be used online for human-robot co-work.

IFAC PAPERSONLINE (2022)

添加到收藏夹

Article Computer Science, Information Systems

Learning Deep Robotic Skills on Riemannian Manifolds

Weitao Wang et al.

Summary: In this paper, the authors propose RiemannianFlow, a deep generative model that enables robots to learn complex and stable skills on Riemannian manifolds. By using distance preserving mappings, the approach transfers data between the original manifold and the tangent space, ensuring the stability of the learning results. Experimental results demonstrate the ability of RiemannianFlow to learn various data patterns and the stability of the learned models.

IEEE ACCESS (2022)

添加到收藏夹

Article Automation & Control Systems

Learning to Assemble Noncylindrical Parts Using Trajectory Learning and Force Tracking

Jianhua Su et al.

Summary: This article introduces a new method for teaching robots assembly skills, combining motion primitive encoding and adaptive impedance control to train trajectory and insertion force simultaneously. Experiments demonstrate the efficiency of this method on noncylindrical parts.

IEEE-ASME TRANSACTIONS ON MECHATRONICS (2022)

添加到收藏夹

Article Robotics

Learning Variable Impedance Control via Inverse Reinforcement Learning for Force-Related Tasks

Xiang Zhang et al.

Summary: This letter proposes an inverse reinforcement learning-based approach to recover both the variable impedance policy and reward function from expert demonstrations, exploring different action spaces of the reward function for a more general representation of expert variable impedance skills. Experiment results showed that the learned reward function in the gain action space has better transferability than in the force space.

IEEE ROBOTICS AND AUTOMATION LETTERS (2021)

添加到收藏夹

Article

Pravila PRISMA 2020.

Matthew J. Page et al.

Medicina Fluminensis (2021)

添加到收藏夹

Article Computer Science, Information Systems

Performance Evaluation of Optical Motion Capture Sensors for Assembly Motion Capturing

Haopeng Hu et al.

Summary: The optical motion capture sensor is widely used in the field of robotics, for tasks such as robot learning from demonstration and 3C product assembly. A performance evaluation protocol is proposed, analyzing the static and trajectory performance of three popular optical MoCaps in 3C product assembly. The experiment results show that all three optical MoCaps meet the requirements of the assembly task, with Prime X41 having the best trajectory performance. This work highlights the potential of optical MoCaps in manufacturing industries.

IEEE ACCESS (2021)

添加到收藏夹

Article Computer Science, Information Systems

Hybrid Trajectory and Force Learning of Complex Assembly Tasks: A Combined Learning Framework

Yan Wang et al.

Summary: The article discusses the need for a precise and adaptive controller for robots to solve complex assembly tasks, which cannot be provided by traditional force control methods. A combined learning-based framework is proposed, utilizing imitation learning and reinforcement learning to address these challenges.

IEEE ACCESS (2021)

添加到收藏夹

Article Robotics

Compliant Parametric Dynamic Movement Primitives

Emre Ugur et al.

ROBOTICA (2020)

添加到收藏夹

Article Automation & Control Systems

Learning peg-in-hole assembly using Cartesian DMPs with feedback mechanism

Nailong Liu et al.

ASSEMBLY AUTOMATION (2020)

添加到收藏夹

Article Robotics

Self-Supervised Learning for Precise Pick-and-Place Without Object Model

Lars Berscheid et al.

IEEE ROBOTICS AND AUTOMATION LETTERS (2020)

添加到收藏夹

Article Robotics

Benchmarking Protocols for Evaluating Small Parts Robotic Assembly Systems

Kenneth Kimble et al.

IEEE ROBOTICS AND AUTOMATION LETTERS (2020)

添加到收藏夹

Article Automation & Control Systems

Robotic Skill Learning for Precision Assembly With Microscopic Vision and Force Feedback

Fangbo Qin et al.

IEEE-ASME TRANSACTIONS ON MECHATRONICS (2019)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

Trajectory generation for robotic assembly operations using learning by demonstration

David A. Duque et al.

ROBOTICS AND COMPUTER-INTEGRATED MANUFACTURING (2019)

添加到收藏夹

Review Business

Literature review as a research methodology: An overview and guidelines

Hannah Snyder

JOURNAL OF BUSINESS RESEARCH (2019)

添加到收藏夹

Article Robotics

Survey of imitation learning for robotic manipulation

Bin Fang et al.

INTERNATIONAL JOURNAL OF INTELLIGENT ROBOTICS AND APPLICATIONS (2019)

添加到收藏夹

Article Robotics

Comparative Peg-in-Hole Testing of a Force-Based Manipulation Controlled Robotic Hand

Karl Van Wyk et al.

IEEE TRANSACTIONS ON ROBOTICS (2018)

添加到收藏夹

Article Automation & Control Systems

Perception of Demonstration for Automatic Programing of Robotic Assembly: Framework, Algorithm, and Validation

Yue Wang et al.

IEEE-ASME TRANSACTIONS ON MECHATRONICS (2018)

添加到收藏夹

Article Automation & Control Systems

Survey on human-robot collaboration in industrial settings: Safety, intuitive interfaces and applications

Valeria Villani et al.

MECHATRONICS (2018)

添加到收藏夹

Article Automation & Control Systems

Robot learning from demonstrations: Emulation learning in environments with moving obstacles

Amir M. E. Ghalamzan et al.

ROBOTICS AND AUTONOMOUS SYSTEMS (2018)

添加到收藏夹

Article Automation & Control Systems

Automated assembly skill acquisition and implementation through human demonstration

Ye Gu et al.

ROBOTICS AND AUTONOMOUS SYSTEMS (2018)

添加到收藏夹

Article Automation & Control Systems

Skill learning and action recognition by arc-length dynamic movement primitives

Timotej Gaspar et al.

ROBOTICS AND AUTONOMOUS SYSTEMS (2018)

添加到收藏夹

Article Automation & Control Systems

Teaching a Robot the Semantics of Assembly Tasks

Thiusius Rajeeth Savarimuthu et al.

IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS (2018)

添加到收藏夹

Review Robotics

Robot Learning from Demonstration in Robotic Assembly: A Survey

Zuyuan Zhu et al.

ROBOTICS (2018)

添加到收藏夹

Article Automation & Control Systems

Optimal Path Planning and Control of Assembly Robots for Hard-Measuring Easy-Deformation Assemblies

An Wan et al.

IEEE-ASME TRANSACTIONS ON MECHATRONICS (2017)

添加到收藏夹

Article Automation & Control Systems

Generalization of orientation trajectories and force-torque profiles for robotic assembly

Aljaz Kramberger et al.

ROBOTICS AND AUTONOMOUS SYSTEMS (2017)

添加到收藏夹

Proceedings Paper Automation & Control Systems

Teaching Assembly by Demonstration using Advanced Human Robot Interaction and a Knowledge Integration Framework

Mathias Haage et al.

27TH INTERNATIONAL CONFERENCE ON FLEXIBLE AUTOMATION AND INTELLIGENT MANUFACTURING, FAIM2017 (2017)

添加到收藏夹

Article Robotics

Survey of robotic manipulation studies intending practical applications in real environments-object recognition, soft robot hand, and challenge program and benchmarking

Tetsuyou Watanabe et al.

ADVANCED ROBOTICS (2017)

添加到收藏夹

Article Computer Science, Theory & Methods

Imitation Learning: A Survey of Learning Methods

Ahmed Hussein et al.

ACM COMPUTING SURVEYS (2017)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

Robot skills for manufacturing: From concept to industrial deployment

Mikkel Rath Pedersen et al.

ROBOTICS AND COMPUTER-INTEGRATED MANUFACTURING (2016)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Adaptation of manipulation skills in physical contact with the environment to reference force profiles

Fares J. Abu-Dakka et al.

AUTONOMOUS ROBOTS (2015)

添加到收藏夹

Article Automation & Control Systems

Benchmarking in Manipulation Research Using the Yale-CMU-Berkeley Object and Model Set

Berk Calli et al.

IEEE ROBOTICS & AUTOMATION MAGAZINE (2015)

添加到收藏夹

Article Robotics

Learning grounded finite-state representations from unstructured demonstrations

Scott Niekum et al.

INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH (2015)

添加到收藏夹

Article Engineering, Industrial

Solving peg-in-hole tasks by human demonstration and exception strategies

Fares J. Abu-Dakka et al.

INDUSTRIAL ROBOT-AN INTERNATIONAL JOURNAL (2014)

添加到收藏夹

Review Business

A Multi-Dimensional Framework of Organizational Innovation: A Systematic Review of the Literature

Mary M. Crossan et al.

JOURNAL OF MANAGEMENT STUDIES (2010)

添加到收藏夹

Review Automation & Control Systems

A survey of robot learning from demonstration

Brenna D. Argall et al.

ROBOTICS AND AUTONOMOUS SYSTEMS (2009)

添加到收藏夹

Article Engineering, Industrial

The impact of the demographic transition on manufacturing Effects of an ageing workforce in German industrial firms

Joern-Henrik Thun et al.

JOURNAL OF MANUFACTURING TECHNOLOGY MANAGEMENT (2007)

添加到收藏夹

© Peeref 2019-2024. All rights reserved.