检索结果-内蒙古大学图书馆

IEEE Conference on Decision and control

作者： Dominik Baumann Krzysztof Kowalczyk Koen Tiels Paweł Wachel Department of Electrical Engineering and Automation Aalto University Espoo Finland Department of Information Technology Uppsala University Uppsala Sweden Department of Control Systems and Mechatronics Wrocław University of Science and Technology Wrocław Poland Department of Mechanical Engineering Eindhoven University of Technology Eindhoven The Netherlands

Safety is an essential asset when learning control policies for physical systems, as violating safety constraints during training can lead to expensive hardware damage. In response to this need, the field of safe learning has emerged with algorithms that can provide probabilistic safety guarantees without knowledge of the underlying system dynamics. Those algorithms often rely on Gaussian process inference. Unfortunately, Gaussian process inference scales cubically with the number of data points, limiting applicability to high-dimensional and embedded systems. In this paper, we propose a safe learning algorithm that provides probabilistic safety guarantees but leverages the Nadaraya-Watson estimator instead of Gaussian processes. For the Nadaraya-Watson estimator, we can reach logarithmic scaling with the number of data points. We provide theoretical guarantees for the estimates, embed them into a safe learning algorithm, and show numerical experiments on a simulated seven-degrees-of-freedom robot manipulator.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Catch me if you can: A pursuit-evasion game with intelligent agents in the Unity 3D game environment 6

Catch me if you can: A pursuit-evasion game with intelligent...

引用

6th International Conference on Electrical engineering, ICEE 2020

作者： Sahin, Ihsan Kumbasar, Tufan Istanbul Technical University Mechatronics Engineering Department Istanbul Turkey Istanbul Technical University Control and Automation Engineering Department Istanbul Turkey

ISBN: (纸本)9781728182926

In this paper, we present a pursuit-evasion game with intelligent agents in the Unity 3D environment. The pursuit-evasion game consists of one evader and four pursuers which are the red, blue, orange and pink agents. In the game scenario, the goal of the evader, which is controlled by human user, is to collect papers without getting caught by the four pursuer agents. The pursuer agents are constructed such that they can move autonomously in the game field via the deployment of a dynamic waypoint structure and the A∗ algorithm. The red, blue and orange pursuer agents share the information on location of the evader agent if the evader agent lies in one of their limited line of sight. They try to intercept the evader agent from different directions in collaborative manner. In this context, we designed a PD controller based pursuing system that tries to minimize the distance between the evader and pursuer agent. For the pink agent, we designed a fuzzy logic based pursuing system that is capable to generate reference interception points and force values to catch evader agent. The performance of the pursuing systems is examined by providing experimental results collected from the game-environment. The results show that the performance of the developed pursuing strategies is satisfactory and the fuzzy logic based pursuing system resulted with a better performance when compared to its conventional counterpart. © 2020 IEEE.

关键词： Fuzzy logic

来源：评论

学校读者我要写书评

暂无评论

Condensed Representation of Machine Learning Data

arXiv

引用

arXiv 2022年

作者： Zengin, Rahman Salim Sezer, Volkan Department of Mechatronics Engineering Istanbul Technical University Istanbul Turkey Department of Control and Automation Engineering Istanbul Technical University Istanbul Turkey Autonomous Mobility Group Electrical and Electronics Engineering Faculty Istanbul Technical University Istanbul Turkey

Training of a Machine Learning model requires sufficient data. The sufficiency of the data is not always about the quantity, but about the relevancy and reduced redundancy. Data-generating processes create massive amounts of data. When used raw, such big data is causing much computational resource utilization. Instead of using the raw data, a proper Condensed Representation can be used instead. Combining K-means, a well-known clustering method, with some correction and refinement facilities a novel Condensed Representation method for Machine Learning applications is introduced. To present the novel method meaningfully and visually, synthetically generated data is employed. It has been shown that by using the condensed representation, instead of the raw data, acceptably accurate model training is possible. Copyright © 2022, The Authors. All rights reserved.

关键词： K-means clustering

来源：评论

学校读者我要写书评

暂无评论

An Improved Cooperative Team Spraying control of a Diffusion Process With a Moving or Static Pollution Source

引用

IEEE/CAA Journal of Automatica Sinica 2020年第2期7卷 494-504页

作者： Juan Chen Baotong Cui Yang Quan Chen Bo Zhuang Key Laboratory of Advanced Process Control for Light Industry(Ministry of Education)and School of IoT Engineering Jiangnan UniversityWuxi 214122China Mechatronics Embedded Systems and Automation LaboratoryUniversity of CaliforniaMerced CA 95343 USA

This paper is concerned with a control problem of a diffusion process with the help of static mesh sensor networks in a certain region of interest and a team of networked mobile actuators carrying chemical *** major contribution of this paper can be divided into three parts:the first is the construction of a cyber-physical system framework based on centroidal Voronoi tessellations(CVTs),the second is the convergence analysis of the actuators location,and the last is a novel proportional integral(PI)control method for actuator motion planning and neutralizing control(e.g.,spraying)of a diffusion process with a moving or static pollution source,which is more effective than a proportional(P)control *** optimal spraying control cost function is ***,the minimization problem of the spraying amount is ***,a new CVT algorithm based on the novel PI control method,henceforth called PI-CVT algorithm,is introduced together with the convergence analysis of the actuators location via a PI control ***,a modified simulation platform called diffusion-mobile-actuators-sensors-2-dimension-proportional integral derivative(Diff-MAS2D-PID)is *** addition,a numerical simulation example for the diffusion process is presented to verify the effectiveness of our proposed controllers.

关键词： Centroidal Voronoi tessellations(CVTs) diffusion processes mobile actuator-sensor networks(MAS-Net) PI control

来源：评论

学校读者我要写书评

暂无评论

A Guideline for Humanoid Leg Design with Oblique Axes for Bipedal Locomotion

A Guideline for Humanoid Leg Design with Oblique Axes for Bi...

引用

IEEE-RAS International Conference on Humanoid Robots

作者： Konrad Fründ Anton Leonhard Shu Florian Christoph Loeffl Christian Ott German Aerospace Center (DLR) Institute of Robotics and Mechatronics Wessling Germany Faculty of Electrical Engineering and Information Technology TU Wien Automation and Control Institute Vienna Austria

ISBN: (纸本)9798350309805

The kinematics of humanoid robots are strongly inspired by the human archetype. A close analysis of the kinematics of the human musculoskeletal system reveals that the human joint axes are oriented within certain inclinations. This is in contrast to the most popular humanoid design with a configuration based on perpendicular joint axes. This paper reviews the oblique joint axes of the mainly involved joints for locomotion of the human musculoskeletal system. We elaborate on how the oblique axes affect the performance of walking and running. The mechanisms are put into perspective for the locomotion types of walking and running. In particular, walking robots can highly benefit from using oblique joint axes. For running, the primary goal is to align the axis of motion to the mainly active sagittal plane. The results of this analysis can serve as a guideline for the kinematic design of a humanoid robot and a prior for optimization-based approaches.

关键词： Legged locomotion Musculoskeletal system Actuators Torque Robot kinematics Humanoid robots Kinematics

来源：评论

学校读者我要写书评

暂无评论

control System for Electric Drives of Plate Straightening Machines

Control System for Electric Drives of Plate Straightening Ma...

引用

International Russian automation Conference (RusAutoCon)

作者： Andrey A. Radionov Andrey Yu. Semitko Boris M. Loginov Vadim R. Khramshin Aleksandra A. Filimonova Department of Automation and Control Moscow Polytechnic University Moscow Russia Department of Mechatronics and Automation South Ural State University Chelyabinsk Russia Magnitogorsk Iron and Steel Works Magnitogorsk Russia Power Engineering and Automated Systems Institute Nosov Magnitogorsk State Technical University Magnitogorsk Russia Department of Electric Drive Mechatronics and Electromechanics South Ural State University Chelyabinsk Russia

The essential quality parameter of rolled products is the flatness along the rolled stock length. The profile defects of the rolled stock, produced at the modern plate mill 5000, frequently occur in the course of intense cooling implied by the thermomechanical rolling technology. To eliminate them, the mill lines include cold straightening machines. The authors show typical flatness defects caused by the occurrence of longitudinal, central and edge waves. Further, the paper characterizes a cold straightening machine installed at the mill 5000 and provides the parameters of electric engines of straightening rollers. The key drawback of the design algorithm for electric drives’ control is setting engine constant moments regardless of the thickness and character of the defects in the rolled stock processed. This causes overloading of the electric and mechanical systems of certain rollers and unsatisfactory product quality. To eliminate the drawback, the authors developed a method for individual setting of engine moments closely coupled in the course of sheet processing. The authors present a functional block diagram for the control system to implement the method. The rationale is provided for the digital setting of moment controllers switched on the inputs of speed controllers. The authors conducted simulation modelling and comparative analysis of engine speeds and moments at the design and proposed settings. The paper provides the analysis of oscillograms obtained before and after the introduction of the control system developed. Its technical and economic feasibility was confirmed during operation.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Utilizing Multi-Agent Deep Reinforcement Learning for Autonomous Intersection Management Systems: A Promising Approach

Utilizing Multi-Agent Deep Reinforcement Learning for Autono...

引用

Artificial Intelligence, Robotics and control (AIRC), International Conference on

作者： Mostafa K. Ghaith Mohamed M. Rehaan N. Shouman Y. Abdalla Omar M. Shehata Automotive and Mechatronics German International University Cairo Egypt Electrical Power Systems German International University Cairo Egypt Automation and Control German International University Cairo Egypt Engineering and Materials Science German University in Cairo Cairo Egypt

In modern cities, intersections are vital pieces of road infrastructure, but they also have the potential to snarl traffic owing to accidents or a lack of traffic coordination systems like traffic signals. As traffic eco-systems become more connected and autonomous, researchers have been increasingly interested in the concept of autonomous intersection management (AIM). AIM improves traffic flow by coordinating the motion of connected vehicles through intersections. This study proposes an AIM system consisting of a two-layer hierarchal control architecture consisting of a vehicle controller layer utilizing Model Predictive control (MPC) that is tuned to maintain the designated trajectory for each vehicle while keeping the passengers comfortable by delivering a smooth acceleration profile. And an Intersection Manager layer utilizing Multi-Agent Reinforcement Learning using centralized training decentralized execution scheme to find the optimal strategy to prevent collisions between vehicles at an intersection by assigning a trajectory for each vehicle entering the intersection. A longitudinal vehicle dynamic model was used to mimic the motion of vehicles in response to the controller’s commands. The proposed AIM system noticeably improved throughput and boosted efficiency without sacrificing safety or passenger comfort.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Development of an Automated Wheelchair for Visually Impaired People 23

Development of an Automated Wheelchair for Visually Impaired...

引用

23rd IEEE International Multi-Topic Conference, INMIC 2020

作者： Siddiqui, Arslan Anwer, Talha Zafar, Saqib Maqbool, Hafiz Farhan Siddiqi, Dilruba Ahmad, Nasir Mehmood, Muhammad Armghan Riaz, Ahsan University of Engineering and Technology Department of Mechatronics and Control Engineering Lahore Pakistan Human Centered Robotics Lab National Center of Robotics and Automation Lahore Pakistan

ISBN: (纸本)9781728198934

This paper presents the development of an automated wheelchair, a mobility-assistive device that helps visually impaired people in the community. For such people, there are many constraints associated while moving to their desired destination. One of the main objectives of the current work is to design and develop a low-cost smart wheelchair that navigates to its destination autonomously. This smart navigation system (SNS) uses a machine vision technique for localization and path planning and proximity sensors for obstacle avoidance. The wheelchair can autonomously navigate to the location selected by the user through a braille keypad interface. The mechanical design is focused to make a finished product that is stable and according to basic medical considerations. The hardware design in the paper aims to enable a visually impaired person to perform daily routine tasks without the need for any assistance. © 2020 IEEE.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

A lightweight pedestrian detection model for edge computing systems 17th

A lightweight pedestrian detection model for edge computing ...

引用

17th International Symposium on Distributed Computing and Artificial Intelligence, DCAI 2020

作者： Chen, Wen-Hui Kuo, Han-Yang Lin, Yu-Chen Tsai, Cheng-Han Graduate Institute of Automation Technology National Taipei University of Technology Taipei Taiwan Department of Automatic Control Engineering Feng Chia University Taichung Taiwan Mechanical and Mechatronics Systems Laboratories Industrial Technology Research Institute Hsinchu Taiwan

ISBN: (纸本)9783030530358

Most vision-based pedestrian detection systems adopted deep learning approaches with the framework of convolutional neural networks (CNNs) to reach the state-of-the-art detection accuracy. As CNN-based approaches are computationally intensive, the deployment of those systems to a resource-limited edge device is a challenging task, especially for a real-time application such as intelligent vehicles. In this study, we proposed a lightweight high-performance edge computing solution to achieve rapid and accurate performance in pedestrian detection. Experimental results showed that the proposed framework can effectively reduce the miss rate of the YOLO-tiny detection model from 48.8% to 26.2% while achieving an inference speed of 31 frames per second (FPS) tested on the Caltech pedestrian dataset. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021.

关键词： Edge computing

来源：评论

学校读者我要写书评

暂无评论

Utilizing the Natural Dynamics of Elastic Legged Robots for Periodic Jumping Motions

Utilizing the Natural Dynamics of Elastic Legged Robots for ...

引用

IEEE-RAS International Conference on Humanoid Robots

作者： Fabian Beck Maximilian Rehermann Johann Reger Christian Ott German Aerospace Center (DLR) Institute of Robotics and Mechatronics Wessling Germany Control Engineering Group Technische Universität Ilmenau Ilmenau Germany Automation and Control Institute Technische Universität Wien Vienna Austria

ISBN: (纸本)9798350309805

This work focuses on the energy efficient control of a planar bipedal robot by using elastic elements in the joints for short-term energy storage. The considered biped exhibits three degrees of freedom per leg and each joint is equipped with a series-elastic actuator (SEA). A controller is developed to enable point foot hopping based on the spring loaded inverted pendulum template. The control design was split into the high-level rigid-body and the elastic dynamics and is validated for hopping motions by numerical simulations. A high -level reference design is proposed, which enables that the elastic system can outperform the corresponding rigid counterpart with regard to energy efficiency in stance phase by more than 50 percent.

关键词： Legged locomotion Technological innovation Dynamics Torque control Humanoid robots Europe Energy efficiency

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：