检索结果-内蒙古大学图书馆

第32届中国控制与决策会议

作者： Yang Jiang Xulong Li Minghao Yu Zhongyu Bai Faculty of Robot Science and Engineering Northeastern University

Manipulating objects is the premise of many robotic applications,such as assembly,sort and *** programming is widely used in industrial robot,it is necessary to improve the intelligent level of *** this paper,we propose an accurate,real-time,light-weight convolutional neural network SqueezeNet-RCM to predict a grasp configuration from a RGB and aligned depth image of graspable *** of using sliding window or regional proposal network to generate potential bounding box,we use a way of end-to-end to train and test our *** the standard Cornell Grasping Dataset,our model achieves accuracy of 90.1% and 88.6% on image-wise split and objectwise split respectively,speed at 73 frames per second(fps) on GPU inferencing,which could meet the requirement of ***,our model size is 2.9 MB,which is able to fit the memory to limited environment such as FPGA.

关键词： Grasp Detection robotic Grasping SqueezeNet-RCM SqueezeNet CNN

来源：评论

学校读者我要写书评

暂无评论

Two-Steam Fully Connected Graph Convolutional Network for Skeleton-Based Action Recognition

Two-Steam Fully Connected Graph Convolutional Network for Sk...

引用

第32届中国控制与决策会议

作者： Zhongyu Bai Qichuan Ding Jiawei Tan Faculty of Robot Science and Engineering Northeastern University

Skeleton-based human action recognition has recently drawn a lot of attentions with the increasing availability of large-scale skeleton *** Convolutional Network(GCN) methods have achieved relatively good performances in action ***,most GCN methods based on predefined graphs with fixed topology constraints always neglect the potential dependencies derived from the cooperative movement of all ***,the lengths and the directions of skeletons are rarely *** easily cause a larger deviation of the estimated action from the actual ***,a two-steam fully connected graph convolutional network(2 s-FGCN) is *** topology structure of the 2 s-FGCN covers the local physical connections and the global potential cooperation of all joints and the joints,lengths and directions of skeletons are all input to the *** experimental results on two datasets(NTU-RGB+D and Kinetics-Skeleton) demonstrate that the proposed model can obtain the state-of-the-art results.

关键词： Action recognition Graph convolutional network(GCN) Skeleton datasets Two-stream framework

来源：评论

学校读者我要写书评

暂无评论

A robot Visual SLAM Algorithm Based on Cloud Computing

A Robot Visual SLAM Algorithm Based on Cloud Computing

引用

作者： Peng Ji Qifeng Liu Hao Chu Faculty of Robot Science and Engineering Northeastern University

Visual SLAM is becoming a hot spot mobile robot navigation. When executing SLAM tasks, traditional robots have many shortcomings, Huge data processing and computing tasks, which consumed much storage and computing resource. There is a high demand for storage and computing resources, which is in contradiction with the limited resources of the airborne hardware. Based on the advantages of cloud computing’s computing resource sharing mechanism, this thesis combines robot with cloud computing to research the mobile robot’s navigation using visual SLAM. Considering that robot visual SLAM is a computationally intensive task, a cloud computing service platform for robot Los SLAM is constructed. The platform is based on Openstack management platform to build a virtual environment, and uses Docker container technology to complete the design of functional service image. The full use of the platform performance makes the application of sophisticated and advanced algorithms in low-cost robots possible. On this basis, an improved visual SLAM algorithm based on ORB features is designed to realize the robot’s effective perception for the unknown environment on the cloud service platform. The results of the algorithm is verified that the service platform can effectively unload the tasks of the robot to the cloud, and reduce the hardware performance requirements of the robot. At the same time, modular functional service forms can expand the functions and use scenarios of robots, break the limits of traditional robots dealing with a single scene, and help to enhance the intelligence of robots.

关键词： Cloud robotics Visual SLAM Cloud service platform ORB features based SLAM Docker container technology

来源：评论

学校读者我要写书评

暂无评论

A Localization Algorithm Based on NLOS Identification with Improved VB-AKF

A Localization Algorithm Based on NLOS Identification with I...

引用

2019人工智能与大数据应用研讨会

作者： Hao Chu Zhibo Fan Fei Wang Peng Ji Chengdong Wu Faculty of Robot Science and Engineering Northeastern University

As a research hotspot, wireless sensor network(WSN) has a wide range of application prospect. The non-line of sight(NLOS) localization is one of the most basic techniques. In order to mitigate the NLOS influence and increase the localization accuracy, we proposes an improved VB-AKF algorithm in this paper. Firstly, the propagation state of the signal is identified. Then, for the LOS and NLOS mix environments, we reduce the NLOS error by changing the attenuation coefficient. And the maximum likelihood(ML) method is used for localization. Finally, the simulation results show that our method can effectively obtain higher localization accuracy.

关键词： Kalman A Localization Algorithm Based on NLOS Identification with Improved VB-AKF

来源：评论

学校读者我要写书评

暂无评论

AI-Based Maize and Weeds Detection on the Edge with CornWeed Dataset

AI-Based Maize and Weeds Detection on the Edge with CornWeed...

引用

Federated Conference on Computer science and Information Systems (FedCSIS)

作者： Naeem Iqbal Christoph Manss Christian Scholz Daniel König Matthias Igelbrink Arno Ruckelshausen Plan-based Robot Control DFKI Osnabrueck Germany Marine Perception DFKI Oldenburg Germany Faculty of Engineering and Computer Science University of Applied Sciences Osnabrueck Osnabrueck Germany

Artificial intelligence (AI) is used more heavily in agricultural applications. Yet, the lack of wireless-fidelity (Wi-Fi) connections on agricultural fields makes AI cloud services unavailable. Consequently, AI models have to be processed directly on the edge. In this paper, we evaluate state-of-the-art detection algorithms for their use in agriculture, in particular plant detection. Thus, this paper presents the CornWeed data set, which has been recorded on farm machines, showing labelled maize crops and weeds for plant detection. The paper provides accuracies for the state-of-the-art detection algorithms on the CornWeed data set, as well as frames per second (FPS) metrics for the considered networks on multiple edge devices. Moreover, for the FPS analysis, the detection algorithms are converted to open neural network exchange (ONNX) and TensoRT engine files as they could be used as future standards for model exchange.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Manipulator Package Sorting and Placing System Based on Computer Vision

Manipulator Package Sorting and Placing System Based on Comp...

引用

第32届中国控制与决策会议

作者： Zhangchao Pan Zixi Jia Kaiyuan Jing Yinguang Ding Qianmeng Liang Faculty of Robot Science and Engineering Northeastern University

Nowadays the logistics industry is developing rapidly. This paper proposes and designs a vision-based manipulator express sorting and delivery system for the problem of express stacking. This system combines computer vision, deep learning, manipulator control and motion planning technologies. This paper uses the YOLO algorithm and PCL to determine the position and normal vector of the pick-up point of the package, plans and controls the movement of the manipulator to the target position to suck the package, and uses vision to determine the size of the package and obtain customer information. Compared with RCNN and SDD algorithm, YOLO has a faster detection speed [1], and at the same time, it handles small stacked targets such as express boxes with higher accuracy. Therefore, this system uses the YOLOv3 algorithm to improve the system’s extraction accuracy and speed. Finally, the success rate of the system was 75%, and the completion time was about 15 s.

关键词： Computer vision Deep learning Size determination Manipulator motion planning

来源：评论

学校读者我要写书评

暂无评论

A Non-line-of-Sight Propagation Identification Algorithm in Wireless Sensor Network

A Non-line-of-Sight Propagation Identification Algorithm in ...

引用

2019人工智能与大数据应用研讨会

作者： H Chu P Ji S Lian H Yang R Wu Faculty of Robot Science and Engineering Northeastern University

With the rapid developing of communication technology, Wireless Sensor Network(WSN) has been applied in many fields. The accuracy of localization is the vital key to all these applications. In order to mitigate the non-line of-sight(NLOS) influence, we propose a propagation identification algorithm based on Anderson-Darling(AD) test. Firstly, the status of signal is determined by the AD testing. Secondly, the particle swarm optimization algorithm is adopted to estimate the mobile node based on the identification results. Finally, the effectiveness of the algorithm is verified by simulation experiments.

关键词： A Non-line-of-Sight Propagation Identification Algorithm in Wireless Sensor Network Sensor

来源：评论

学校读者我要写书评

暂无评论

Adaptive dynamic environment response based evolutionary algorithm for dynamic multi-objective optimization

Adaptive dynamic environment response based evolutionary alg...

引用

第34届中国控制与决策会议

作者： Kanrong Liu Jianchang Liu Shubin Tan Fei Li Tianzi Zheng Yuanchao Liu School of Information Science and EngineeringNortheastern University Department of Electrical and Information Engineering Anhui University of Technology Anhui Province Key Laboratory of Special Heavy Load Robot Anhui Unviersity of Technology

In dynamic multi-objective optimization problems(DMOPs),multiple conflicting objectives vary over ***,a core issue for the DMOPs is to respond the dynamic environments,when the environment change has been *** effectively address DMOPs,a novel method,called adaptive dynamic environment response based dynamic multi-objective evolutionary algorithm(ADER-DMOEA),is proposed in this *** proposed algorithm can adaptively respond the dynamic environment based on the detected environment change ***,a novel environment change severity detection mechanism is developed,where C-metric is introduced to measure the performance of different population obtained by different types of response ***,according to the measurement results,the algorithm adaptively adopts different strategies for environment ***,it can ensure that the proposed method has a good initial population for each new *** proposed method is compared with three algorithms,and the experiment results have demonstrated that ADER-DMOEA is competitive in dealing with dynamic environments.

关键词： Adaptive dynamic environment response Dynamic multi-objective optimization Prediction Memory Response mechanism C-metric

来源：评论

学校读者我要写书评

暂无评论

Exploiting More Information in Sparse Point Cloud for 3D Single Object Tracking

arXiv

引用

arXiv 2022年

作者： Cui, Yubo Shan, Jiayao Gu, Zuoxu Li, Zhiheng Fang, Zheng The Faculty of Robot Science and Engineering Northeastern University Shenyang110819 China The Science and Technology Near-Surface Detection Laboratory Wuxi214000 China The National Frontiers Science Center for Industrial Intelligence and Systems Optimization Shenyang110819 China The The Key Laboratory of Data Analytics and Optimization for Smart Industry Northeastern University Ministry of Education Shenyang110819 China

3D single object tracking is a key task in 3D computer vision. However, the sparsity of point clouds makes it difficult to compute the similarity and locate the object, posing big challenges to the 3D tracker. Previous works tried to solve the problem and improved the tracking performance in some common scenarios, but they usually failed in some extreme sparse scenarios, such as for tracking objects at long distances or partially occluded. To address the above problems, in this paper, we propose a sparse-to-dense and transformer-based framework for 3D single object tracking. First, we transform the 3D sparse points into 3D pillars and then compress them into 2D bird's eye view (BEV) features to have a dense representation. Then, we propose an attention-based encoder to achieve global similarity computation between template and search branches, which could alleviate the influence of sparsity. Meanwhile, the encoder applies the attention on multi-scale features to compensate for the lack of information caused by the sparsity of point cloud and the single scale of features. Finally, we use set-prediction to track the object through a two-stage decoder which also utilizes attention. Extensive experiments show that our method achieves very promising results on the KITTI and NuScenes datasets. Code is available at https://***/3bobo/smat. © 2022, CC BY.

关键词： Signal encoding

来源：评论

学校读者我要写书评

暂无评论

Dual-Objective Mixed Integer Linear Program and Memetic Algorithm for an Industrial Group Scheduling Problem

引用

IEEE/CAA Journal of Automatica Sinica 2021年第6期8卷 1199-1209页

作者： Ziyan Zhao Shixin Liu MengChu Zhou Abdullah Abusorrah State Key Laboratory of Synthetical Automation for Process Industries and the College of Information Science and Engineering Northeastern UniversityShenyang 110819China Department of Electrical and Computer Engineering New Jersey Institute of TechnologyNewark NJ 07102 USA Center of Research Excellence in Renewable Energy and Power Systems King Abdulaziz UniversityJeddah 21481Saudi Arabia Department of Electrical and Computer Engineering Faculty of Engineeringand Center of Research Excellence in Renewable Energy and Power SystemsKing Abdulaziz UniversityJeddah 21481Saudi Arabia

Group scheduling problems have attracted much attention owing to their many practical *** work proposes a new bi-objective serial-batch group scheduling problem considering the constraints of sequence-dependent setup time,release time,and due *** is originated from an important industrial process,i.e.,wire rod and bar rolling process in steel production *** objective functions,i.e.,the number of late jobs and total setup time,are minimized.A mixed integer linear program is established to describe the *** obtain its Pareto solutions,we present a memetic algorithm that integrates a population-based nondominated sorting genetic algorithm II and two single-solution-based improvement methods,i.e.,an insertion-based local search and an iterated greedy *** computational results on extensive industrial data with the scale of a one-week schedule show that the proposed algorithm has great performance in solving the concerned problem and outperforms its *** high accuracy and efficiency imply its great potential to be applied to solve industrial-size group scheduling problems.

关键词： Insertion-based local search iterated greedy algorithm machine learning memetic algorithm nondominated sorting genetic algorithm II(NSGA-II) production scheduling

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：