检索结果-内蒙古大学图书馆

Inertial Estimation and Energy-Efficient Control of a Cable-suspended Load with a Team of UAVs

学校读者我要写书评

暂无评论

Inertial Estimation and Energy-Efficient Control of a Cable-...

International Conference on Unmanned Aircraft systems (ICUAS)

作者： A. Petitti D. Sanalitro M. Tognon A. Milella J. Cortés A. Franchi Institute of Intelligent Industrial Systems and Technologies for Advanced Manufacturing (STIIMA) CNR Bari Italy LAAS-CNRS Université de Toulouse CNRS Toulouse France Autonomous Systems Lab ETH Zurich Zürich Switzerland Robotics and Mechatronics lab University of Twente Enschede The Netherlands

ISBN: (数字)9781728142784

ISBN: (纸本)9781728142791

The Fly-Crane is a multi-robot aerial manipulator system composed of three aerial vehicles towed to a platform by means of six cables. This paper presents a method to estimate the mass and the position of the center of mass of a loaded platform (i.e. the Fly-Crane platform including a transported load). The precise knowledge of these parameters allows to sensibly minimize the total effort exerted during a full-pose manipulation task The estimation is based on the measure of the forces applied by the aerial vehicles to the platform in different static configurations. We demonstrate that only two different configurations are sufficient to estimate the inertial parameters. Far-from-ideal numerical simulations show the effectiveness of the estimation method. Once the parameters are estimated, we show the enhancement of the system performances by minimizing the total exerted effort. The validity of the proposed algorithm in non-ideal conditions is presented through simulations based on the Gazebo simulator.

关键词： Estimation Task analysis Numerical models Robots Vehicle dynamics Trajectory Load modeling

Encrypted Feedback Linearization and Motion Control for Manipulator with Somewhat Homomorphic Encryption

学校读者我要写书评

暂无评论

Encrypted Feedback Linearization and Motion Control for Mani...

IEEE/ASME (AIM) International Conference on Advanced intelligent Mechatronics

作者： Kaoru Teranishi Kiminao Kogiso Jun Ueda Department of Mechanical Engineering and Intelligent Systems The University of Electro-Communications Tokyo Chofu Japan Bio-Robotics & Human Modeling Lab School of Mechanical Engineering Georgia Institute of Technology Atlanta GA USA

ISBN: (数字)9781728167947

ISBN: (纸本)9781728167954

Encrypted control is a cryptographic approach to reducing the risks of eavesdropping attacks for networked control systems. An encrypted controller determines control inputs directly from encrypted data without decryption. This study considers an implementation of an encrypted motion controller, which consists of feedback linearization and PD control, for a manipulator using somewhat homomorphic encryption. A controller reconstruction method is proposed for separating a nonlinear time-varying controller into a linear part and a nonlinear part. A plant side device supports encrypted controller operations by generating the nonlinear part in advance of encryption. Simulation results demonstrate that the proposed encrypted control system can achieve the same control performance as the one that an unencrypted control system does.

关键词： Encryption Manipulators Networked control systems Quantization (signal) Feedback linearization

MoGaze: A dataset of full-body motions that includes workspace geometry and eye-gaze

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Kratzer, Philipp Bihlmaier, Simon Midlagajni, Niteesh Balachandra Prakash, Rohit Toussaint, Marc Mainprice, Jim Machine Learning and Robotics Lab University of Stuttgart Humans to Robots Motions Research Group University of Stuttgart Germany Humans to Robots Motions Research Group University of Stuttgart Germany Learning and Intelligent Systems lab TU Berlin Germany

As robots become more present in open human environments, it will become crucial for robotic systems to understand and predict human motion. Such capabilities depend heavily on the quality and availability of motion capture data. However, existing datasets of full-body motion rarely include 1) long sequences of manipulation tasks, 2) the 3D model of the workspace geometry, and 3) eye-gaze, which are all important when a robot needs to predict the movements of humans in close proximity. Hence, in this paper, we present a novel dataset of full-body motion for everyday manipulation tasks, which includes the above. The motion data was captured using a traditional motion capture system based on reflective markers. We additionally captured eye-gaze using a wearable pupil-tracking device. As we show in experiments, the dataset can be used for the design and evaluation of full-body motion prediction algorithms. Furthermore, our experiments show eye-gaze as a powerful predictor of human intent. The dataset includes 180 min of motion capture data with 1627 pick and place actions being performed. It is available at MoGaze Dataset1 and is planned to be extended to collaborative tasks with two humans in the near future. Copyright © 2020, The Authors. All rights reserved.

关键词： Motion estimation

Collaborative evolutionary reinforcement learning

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Khadka, Shauharda Majumdar, Somdeb Nassar, Tarek Dwiel, Zach Tumer, Evren Miret, Santiago Liu, Yinyin Tumer, Kagan Intel AI Lab Collaborative Robotics and Intelligent Systems Institute Oregon State University

Deep reinforcement learning algorithms have been successfully applied to a range of challenging control tasks. However, these methods typically struggle with achieving effective exploration and are extremely sensitive to the choice of hyperparameters. One reason is that most approaches use a noisy version of their operating policy to explore-thereby limiting the range of exploration. In this paper, we introduce Collaborative Evolutionary Reinforcement Learning (CERL), a scalable framework that comprises a portfolio of policies that simultaneously explore and exploit diverse regions of the solution space. A collection of learners-typically proven algorithms like TD3-optimize over varying time-horizons leading to this diverse portfolio. All learners contribute to and use a shared replay buffer to achieve greater sample efficiency. Computational resources are dynamically distributed to favor the best learners as a form of online algorithm selection. Neuroevolution binds this entire process to generate a single emergent learner that exceeds the capabilities of any individual learner. Experiments in a range of continuous control benchmarks demonstrate that the emergent learner significantly outperforms its composite learners while remaining overall more sample-efficient-notably solving the Mujoco Humanoid benchmark where all of its composite learners (TD3) fail entirely in isolation. Copyright © 2019, The Authors. All rights reserved.

关键词： Deep learning

A novel multi-scale fourier descriptor based on plane orthogonal vectors for fast shape retrieval

学校读者我要写书评

暂无评论

A novel multi-scale fourier descriptor based on plane orthog...

2019 IEEE International Conference on robotics and Biomimetics, ROBIO 2019

作者： Ma, Jiajun Zhu, Jiafeng Liang, Guoyuan Wu, Xinyu Guangdong Provincial Key Lab of Robotics and Intelligent System CAS Key Laboratory of Human-Machine Intelligence-Synergy Systems Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Shenzhen Guangdong Province China

ISBN: (纸本)9781728163215

Shape retrieval is an interesting and challenging issue in computer vision and robot vision. The recent works employing Fourier-based multi-scale descriptors achieve some success in shape retrieval, especially make real-time application possible by improving the efficiency. However, these methods are commonly adopting really complex shape descriptors to present the shape information which is not intuitive enough. In our work, we propose a new Fourier-based method with novel multi-scale descriptor which is a combination of contour feature and regional feature. This novel descriptor has two main characteristics: First, utilize six different invariant signatures which are based on plane orthogonal vectors;Second, leverage multiple scale structures to describe local and global information of shapes. Except the easiness in understanding, this descriptor approach also produces a more distinct topology of shapes than other approaches which means the property of stability on variations, linear or nonlinear. The method is tested on some popular databases that provides a promising retrieval rate(87.46% and 99.27%) with constant time complexity optimization. The results confirm that our method is more capable of representing the topology of shapes than most other methods based on Fourier descriptor. © 2019 IEEE.

关键词： Topology

Learn and Transfer Knowledge of Preferred Assistance Strategies in Semi-Autonomous Telemanipulation

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Tao, Lingfeng Bowman, Michael Zhou, Xu Zhang, Jiucai Zhang, Xiaoli Colorado School of Mines Intelligent Robotics and Systems Lab 1500 Illinois St GoldenCO80401 United States The GAC R&D Center Silicon Valley SunnyvaleCA94085 United States

Enabling robots to provide effective assistance yet still accommodating the operator's commands for telemanipulation of an object is very challenging because robot's assistive action is not always intuitive for human operators and human behaviors and preferences are sometimes ambiguous for the robot to interpret. Although various assistance approaches are being developed to improve the control quality from different optimization perspectives, the problem still remains in determining the appropriate approach that satisfies the fine motion constraints for the telemanipulation task and preference of the operator. To address these problems, we developed a novel preference-aware assistance knowledge learning approach. An assistance preference model learns what assistance is preferred by a human, and a stagewise model updating method ensures the learning stability while dealing with the ambiguity of human preference data. Such a preference-aware assistance knowledge enables a teleoperated robot hand to provide more active yet preferred assistance toward manipulation success. We also developed knowledge transfer methods to transfer the preference knowledge across different robot hand structures to avoid extensive robot-specific training. Experiments to telemanipulate a 3-finger hand and 2-finger hand, respectively, to use, move, and hand over a cup have been conducted. Results demonstrated that the methods enabled the robots to effectively learn the preference knowledge and allowed knowledge transfer between robots with less training effort. Copyright © 2020, The Authors. All rights reserved.

关键词： Behavioral research

Forming Real-World Human-Robot Cooperation for Tasks with General Goal

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Tao, Lingfeng Bowman, Michael Zhang, Jiucai Zhang, Xiaoli Colorado School of Mines Intelligent Robotics and Systems Lab 1500 Illinois St GoldenCO80401 United States The GAC R&D Center Silicon Valley SunnyvaleCA94085 United States

In Human-Robot Cooperation, the robot cooperates with humans to accomplish the task together. Existing approaches assume the human has a specific goal during the cooperation, and the robot infers and acts toward it. However, in real-world environments, a human usually only has a general goal (e.g., general direction or area in motion planning) at the beginning of the cooperation, which needs to be clarified to a specific goal (e.g., an exact position) during cooperation. The specification process is interactive and dynamic, which depends on the environment and the partners’ behavior. The robot that does not consider the goal specification process may cause frustration to the human partner, elongate the time to come to an agreement, and compromise or fail team performance. This work presents the Evolutionary Value Learning approach to model the dynamics of the goal specification process with State-based Multivariate Bayesian Inference and goal specificity-related features. This model enables the robot to enhance the process of human’s goal specification actively and find a cooperative policy in a Deep Reinforcement Learning manner. Our method outperforms existing methods with faster goal specification processes and better team performance in a dynamic ball balancing task with real human subjects. Copyright © 2020, The Authors. All rights reserved.

关键词： Reinforcement learning

Learn Task First or Learn Human Partner First: A Hierarchical Task Decomposition Method for Human-Robot Cooperation

学校读者我要写书评

暂无评论

arXiv 2020年

Applying Deep Reinforcement Learning (DRL) to Human-Robot Cooperation (HRC) in dynamic control problems is promising yet challenging as the robot needs to learn the dynamics of the controlled system and dynamics of the human partner. In existing research, the robot powered by DRL adopts coupled observation of the environment and the human partner to learn both dynamics simultaneously. However, such a learning strategy is limited in terms of learning efficiency and team performance. This work proposes a novel task decomposition method with a hierarchical reward mechanism that enables the robot to learn the hierarchical dynamic control task separately from learning the human partner's behavior. The method is validated with a hierarchical control task in a simulated environment with human subject experiments. Our method also provides insight into the design of the learning strategy for HRC. The results show that the robot should learn the task first to achieve higher team performance and learn the human first to achieve higher learning efficiency. Copyright © 2020, The Authors. All rights reserved.

关键词： Deep learning

Improved dynamic window approach with ellipse equations for autonomous navigation of unmanned surface vehicle

学校读者我要写书评

暂无评论

Journal of Institute of Control, robotics and systems 2020年第8期26卷 624-629页

作者： Kim, Hyogon Yun, Sung-Jo Choi, Young-Ho Ryu, Jae-Kwan Won, Byong-Jae Suh, Jin-Ho Korea Institute of Robotics and Technology Convergence Korea Republic of Unmanned/Intelligent Robotic Systems R&D Lab. LIG Nex1 Korea Republic of Department. of Mechanical System Engineering Pukyoung National University Korea Republic of

Autonomous navigation technology for Unmanned Surface Vehicles (USVs) has seen many recent advancements. In order for a USV to operate autonomously, the path to move from its current location to its destination must be planned, and local path planning must be applied for tracking the path and avoiding obstacles. The Dynamic Window Approach (DWA) is a well-known navigation algorithm used for local path planning. The DWA algorithm derives the linear velocity and angular velocity by evaluating the object function defined as destination direction, velocity, and distance from the obstacle. However, because DWA does not consider the velocity and direction of obstacles, USVs may be in danger when avoiding high-speed obstacles. In this paper, we propose an improved DWA algorithm that considers the velocity and direction of obstacles. Whereas the existing DWA algorithm defines the obstacle recognition area as a circle, the proposed DWA algorithm defines the obstacle recognition area as an ellipse and reflects the speed of the obstacle in the ellipse. In this study, we implemented a simulation to compare the existing DWA algorithm with the improved DWA algorithm proposed here. The results confirm that the proposed DWA algorithm can avoid dynamic obstacles more safely than the existing DWA algorithm. © ICROS 2020.

关键词： Unmanned surface vehicles