检索结果-内蒙古大学图书馆

2018 IEEE Intelligent Vehicles Symposium, IV 2018

作者： Xing, Yang Tang, Jianlin Liu, Hong Lv, Chen Cao, Dongpu Velenis, Efstathios Wang, Fei-Yue Advanced Vehicle Engineering Centre Cranfield University BedfordMK43 0AL United Kingdom Vehicle Intelligence Pioneers Ltd Qingdao China Jiangsu XCMG Construction Machinery Research Institute Ltd. No. 26 Tuolanshan Road Economic Development Zone Xuzhou Jiangsu221004 China School of Mechanical and Mechatronics Engineering University of Waterloo 200 University Avenue West WaterlooONN2L 3G1 Canada State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China

ISBN: (纸本)9781538644522

Drivers' decision and their corresponding behaviors are important aspects that can affect the driving safety, and it is necessary to understand the driver behaviors in real-time. In this study, an end-to-end driving-related tasks recognition system is proposed. Specifically, seven common driving activities are identified, which are normal driving, right mirror checking, rear mirror checking, left mirror checking, using in-vehicle video device, texting, and answering mobile phone. Among these, the first four activities are regarded as normal driving tasks, while the rest three are divided into distraction group. The images are collected using a consumer range camera, namely, Kinect. In total, five drivers are involved in the naturalistic data collection. Before training the identification model, the raw images are first segmented using a Gaussian mixture model (GMM) to extract the driver region from the background. Then, a pre-trained deep convolutional neural network (CNN) model is trained to classify the behaviors, which directly takes the processed RGB images as the input and outputs the identified label. In this work, the AIexNet is selected as the pre-trained CNN model. Then, to reduce the training cost, the transfer learning mechanism is applied to the CNN model. An average of 79\% detection accuracy is achieved for the seven driving tasks. The proposed integration model can be used as a low-cost driver distraction and dangerous tasks recognition modeL. © 2018 IEEE.

关键词： Neural network models

来源：评论

学校读者我要写书评

暂无评论

Dual refinement networks for accurate and fast object detection in real-world scenes

arXiv

引用

arXiv 2018年

作者： Chen, Xingyu Yu, Junzhi Kong, Shihan Wu, Zhengxing Wen, Li State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 University of Chinese Academy of Sciences Beijing100049 China Beijing Innovation Center for Engineering Science and Advanced Technology Peking University Beijing100871 China School of Mechanical Engineering and Automation Beihang University Beijing100191 China

Object detection has been vigorously studied for years but fast accurate detection for real-world scenes remains a very challenging problem. Overcoming drawbacks of singlestage detectors, we take aim at precisely detecting objects for static and temporal scenes in real time. Firstly, as a dual refinement mechanism, a novel anchor-offset detection including an anchor refinement, a feature offset refinement, and a deformable detection head is designed for two-step regression and capture of accurate detection features. Based on the anchoroffset detection, a dual refinement network (DRN) is developed for high-performance static detection, where a multi-deformable head is further designed to leverage contextual information for describing objects. As for temporal detection in real-world scenes, temporal refinement networks (TRN) and temporal dual refinement networks (TDRN) are developed by propagating the refinement information across time, where we also propose a loose refinement strategy to match object motion with the previous refinement. Our proposed methods are evaluated on PASCAL VOC, COCO, and ImageNet VID datasets. Extensive comparison on static and temporal detection verify the superiority of DRN, TRN, and TDRN. Consequently, our developed approaches run in a fairly fast speed, and in the meantime achieve a significantly enhanced detection accuracy, i.e., a mAP of 82.8% on VOC 2007, 80.6% on VOC 2012, 69.4% on VID 2017, and an AP of 34.3% on COCO. Ultimately, producing encouraging results, our methods are applied to underwater object grasping with an autonomous system. Codes are publicly available at http://***/SeanChenxy/TDRN. Copyright © 2018, The Authors. All rights reserved.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

Frequency estimation of a sinusoidal signal with time-varying amplitude*

引用

IFAC-PapersOnLine 2017年第1期50卷 12880-12885页

作者： Vedyakov, Alexey A. Vediakova, Anastasiia O. Bobtsov, Alexey A. Pyrkin, Anton A. Aranovskiy, Stanislav V. Department of Control Systems and Informatics ITMO University Kronverkskiy pr. 49 St. Petersburg197101 Russia Department of Computer Applications and Systems St.Petersburg State University 7/9 Universitetskaya nab. St. Petersburg199034 Russia Laboratory “Control of Complex Systems” Institute of Problems of Mechanical Engineering V.O. Bolshoj pr. 61 St. Petersburg199178 Russia CentraleSupélec IETR Avenue de la Boulaie Cesson-Sévigné35576 France

This paper is devoted to frequency estimation of a sinusoidal signal with a time-varying amplitude. The amplitude can take zero values, but should be known and bounded. The first order linear regression model is obtained. The estimation algorithm has been constructed basing on standard gradient approach. It is shown that the frequency estimation error converges to zero exponentially fast. The described algorithm does not require measuring or calculating derivatives of the input signal. The efficiency of the proposed approach is demonstrated through a set of numerical simulations. © 2017

关键词： Frequency estimation Continuous time systems Linear regression Continuous time estimations Estimation algorithm Frequency estimation errors Gradient approach Gray box models Linear regression models Sinusoidal signals Time domain identification

来源：评论

学校读者我要写书评

暂无评论

Adaptive Tracking of a Multi-Sinusoidal Signal with DREM-Based Parameters Estimation

引用

IFAC-PapersOnLine 2017年第1期50卷 4282-4287页

作者： Borisov O.I. Gromov V.S. Vedyakov A.A. Pyrkin A.A. Bobtsov A.A. Aranovskiy S.V. Department of Control Systems and Informatics ITMO University Kronverkskiy pr. 49 St. Petersburg 197101 Russian Federation Laboratory “Control of Complex Systems” Institute of Problems of Mechanical Engineering V.O. Bolshoj pr. 61 St. Petersburg 199178 Russian Federation CentraleSupélec - IETR Avenue de la Boulaie Cesson-Sévigné 35576 France

In this paper tracking of an unknown multi-sinusoidal signal with fast online frequency estimation is addressed. The latter has been achieved by the so-called dynamic regressor extension and mixing (DREM) approach, which allows to increase performance of multiple frequency estimation. Designed predictive compensation algorithm is capable to operate under presence of computational or transmission delays. The resultant controller has been implemented to a robotic arm with a computer vision system. An extensive experimental study shows undoubted advantages of the DREM-based frequency estimator in comparison to the classical gradient approach. A detailed analysis of the transient response properties is provided in the paper. © 2017

关键词： Adaptive control frequency estimation robotic manipulators target tracking time delay

来源：评论

学校读者我要写书评

暂无评论

Reinforcement Learning-Based Predictive control for Autonomous Electrified Vehicles

Reinforcement Learning-Based Predictive Control for Autonomo...

引用

IEEE Symposium on Intelligent Vehicle

作者： Teng Liu Chao Yang Chuanzheng Hu Hong Wang Li Li Dongpu Cao Fei-Yue Wang Vehicle Intelligence Pioneers Inc. Qingdao Shandong China Jiangsu XCMG Construction Machinery Research Institute Ltd. Xuzhou Jiangsu China Mechanical and Mechatronics Engineering Department Waterloo University Canada Faculty of Department of Automation Tsinghua University China State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing China

This paper proposes a learning-based predictive control technique for self-driving hybrid electric vehicle (HEV). This approach is a hierarchical framework. The higher-level is a human-like driver model, which is applied to predict accelerations in the car following situation to replicate a human driver's demonstrations. The lower-level is a reinforcement learning (RL)-based controller, which enforces the battery and fuel consumption constraints to improve energy efficiency of HEV. In addition, we present induced matrix norm (IMN) to handle cases that the training data cannot provide sufficient information on how to operate in current driving situation. Simulation results illustrate that the proposed method can reproduce human driver's driving style and promote fuel economy.

关键词： Acceleration Hybrid electric vehicles Trajectory Hidden Markov models Predictive control Energy efficiency

来源：评论

学校读者我要写书评

暂无评论

The DREM Approach for Chaotic Oscillators Parameter Estimation with Improved Performance

引用

IFAC-PapersOnLine 2017年第1期50卷 7027-7031页

作者： Gromov V.S. Borisov O.I. Pyrkin A.A. Bobtsov A.A. Kolyubin S.A. Aranovskiy S.V. Department of Control Systems and Informatics ITMO University Kronverkskiy pr. 49 St. Petersburg 197101 Russian Federation Laboratory “Control of Complex Systems” Institute of Problems of Mechanical Engineering V.O. Bolshoj pr. 61 St. Petersburg 199178 Russian Federation CentraleSupélec - IETR Avenue de la Boulaie Cesson-Sévigné 35576 France

In this paper we address the problem of parameter estimation performance enhancement for Duffing-like chaotic oscillator with parametric uncertainties. The Dynamic Regressor Extension and Mixing (DREM) approach, which has been recently proposed by the authors, is proven to be a powerful tool in transients improvement; however, it was never applied before for chaotic systems. Here we elaborate on how the DREM approach can be applied for the parameter reconstruction problem of chaotic signal generator used in the data transmitting scenario. The set of numerical examples illustrates the benefits of the proposed approach. © 2017

关键词： Chaotic behavior Observers Transmitters

来源：评论

学校读者我要写书评

暂无评论

Output Robust control with Anti-Windup Compensation for Quadcopters

引用

IFAC-PapersOnLine 2016年第13期49卷 287-292页

作者： Borisov, Oleg I. Gromov, Vladislav S. Pyrkin, Anton A. Bobtsov, Alexey A. Nikolaev, Nikolay A. Department of Control Systems and Informatics ITMO University Kronverkskiy pr. 49 St. Petersburg 197101 Russia Laboratory “Control of Complex Systems” Institute for Problems of Mechanical Engineering Bolshoy pr. 61 St. Petersburg 199178 Russia

In the paper an output control approach for quad copters under the condition of the bounded input signal is presented. This algorithm is based on the high-gain principle "consecutive compensator", which was augmented by an auxiliary integral loop in order to implement the anti-windup scheme. The mathematical model describing quadcopters is decomposed on two parts: a static MIMO transformation and six SISO dynamical channels. The controller generates virtual input signals for these channels, which after inverse MIMO transformation are allocated between the actuators as real bounded control signals. © 2016

关键词： Mathematical transformations MIMO systems Robust control Anti windup compensation Anti windup schemes Bounded controls Bounded input control applications Output control Output regulation Saturation control

来源：评论

学校读者我要写书评

暂无评论

Robotic Boat Setup for control Research and Education

引用

IFAC-PapersOnLine 2016年第6期49卷 256-261页

作者： Borisov, Oleg I Gromov, Vladislav S Pyrkin, Anton A Bobtsov, Alexey A Nikolaev, Nikolay A Department of Control Systems and Informatics ITMO University Kronverkskiy pr. 49 St. Petersburg197101 Russia Laboratory “Control of Complex Systems” Institute for Problems of Mechanical Engineering Bolshoy pr. 61 St. Petersburg199178 Russia

In this paper a robotic boat setup for research and education in control is presented. It was designed for students to develop their practical experience and validate theoretical results at the course “control Methods for Robotic Applications”, which is being taught at the Department of control systems and Informatics of ITMO University. Four use cases of this setup are suggested in the paper. They are aimed to design controllers of various structures. © 2016

关键词： Robotics Boats control systems control applications control education control methods Design controllers Informatics Practical experience Robotic applications Ship control

来源：评论

学校读者我要写书评

暂无评论

Manipulation Tasks in Robotics Education

引用

IFAC-PapersOnLine 2016年第6期49卷 22-27页

作者： Borisov, Oleg I Gromov, Vladislav S Pyrkin, Anton A Vedyakov, Alexey A Petranevsky, Igor V Bobtsov, Alexey A Salikhov, Vladimir I Department of Control Systems and Informatics ITMO University Kronverkskiy pr. 49 St. Petersburg197101 Russia Laboratory “Control of Complex Systems” Institute for Problems of Mechanical Engineering Bolshoy pr. 61 St. Petersburg199178 Russia

In this paper a series of practical tasks on control for manipulators is presented. This series represents a laboratory part of the course “control Methods for Robotic Applications” included in the curriculum of the master's degree program at the Department of control systems and Informatics of ITMO University. The exercises are aimed to attract and motivate students, improve their practical competency and simplify their understanding of theoretical background. The tasks are focused on such problems as path and trajectory planning, automatic code generation and design of a tracking system for moving target. © 2016

关键词： Manipulators Application programs Automatic programming Robotics Automatic code generations control applications control education Manipulation task Robotic applications Robotic manipulators Robotics education Trajectory Planning

来源：评论

学校读者我要写书评

暂无评论

Arc Approximation Algorithm of Spatial Movements for Industrial Robots

Arc Approximation Algorithm of Spatial Movements for Industr...

引用

Annual Conference of the IEEE Industrial Electronics Society

作者： Oleg I. Borisov Vladislav S. Gromov Anton A. Pyrkin Igor V. Petranevsky Alexcy O. Klyunin Alexcy A. Bobtsov Department of Control Systems and Informatics ITMO University St. Petersburg Russia Institute of Automation Hangzhou Dianzi University Hangzhou Zhejiang Province P. R. China Laboratory “Control of Complex Systems” Institute of Problems of Mechanical Engineering St. Petersburg Russia

A new path planning algorithm for industrial robots is proposed in this paper. The algorithm is based on arc approximation of the given reference path. Such solution allows to reduce amount of the reference points, decrease code size and computational costs, increase performance of operations and simplify programming complex movements. The study represents extension of the previous work to the case of planning spatial movements in three-dimensional space. Experiments using a 6-DOF (degrees of freedom) articulated robot simulating typical industrial operations (e.g. welding, engraving, polishing) were carried out and corresponding experimental results are reported in the paper.

关键词： Service robots Planning Approximation algorithms Robot kinematics Welding Software

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：