2021 International Conference on Computers and Automation (CompAuto)最新文献

On-Demand Trajectory Prediction Based on Adaptive Interaction Car Following Model with Decreasing Tolerance 基于递减容差自适应交互跟随模型的按需轨迹预测

2021 International Conference on Computers and Automation (CompAuto) Pub Date : 2021-09-01 DOI: 10.1109/CompAuto54408.2021.00020

Zhanhui Zhou, Jiawen Wu, Zhuo Cao, Zhangcong She, Jiahuai Ma, Xuan‐Wen Zu

{"title":"On-Demand Trajectory Prediction Based on Adaptive Interaction Car Following Model with Decreasing Tolerance","authors":"Zhanhui Zhou, Jiawen Wu, Zhuo Cao, Zhangcong She, Jiahuai Ma, Xuan‐Wen Zu","doi":"10.1109/CompAuto54408.2021.00020","DOIUrl":"https://doi.org/10.1109/CompAuto54408.2021.00020","url":null,"abstract":"Car following is a simple but important scenario when predicting the car trajectory. Deep neural network is good at modeling such interactions between cars, but it is susceptible to noises in the observation data especially when the observation window is short, and overfitting occurs easily even if it is pretrained to get good initial weights in the first place. The same problem exists in model-based methods, these methods are either poor at adaptability or easy to overfit. Therefore, in our research we tackle these problems by introducing our adaptive interaction model with decreasing tolerance. It is a model-based method utilizing both offline training on dataset and online adaptation on observations. Decreasing tolerance makes sure that the model filters out outliers while trying to keep enough reasonable observation data to adapt to. Decreasing tolerance and the combination of offline training with online adaptation together helps to strike a balance between generalization and adaptability in predicting future trajectory. Our code base can be found in https://github.com/ZHZisZZ/AutoDrive-NGSIM.","PeriodicalId":236754,"journal":{"name":"2021 International Conference on Computers and Automation (CompAuto)","volume":"327 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115768684","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Overlapping Community Detection Based on Node Correlation in Directed Complex Networks 有向复杂网络中基于节点关联的重叠社团检测

2021 International Conference on Computers and Automation (CompAuto) Pub Date : 2021-09-01 DOI: 10.1109/CompAuto54408.2021.00018

Junqiang Liu, Chenxi Ma, Yufan Liao, Li Liu

{"title":"Overlapping Community Detection Based on Node Correlation in Directed Complex Networks","authors":"Junqiang Liu, Chenxi Ma, Yufan Liao, Li Liu","doi":"10.1109/CompAuto54408.2021.00018","DOIUrl":"https://doi.org/10.1109/CompAuto54408.2021.00018","url":null,"abstract":"With the rapid development of science and technology, there are all kinds of flow-based networks in our life, such as fraudulent fund transfer networks, etc. How to use the community detection algorithms to effectively detect the communities contained in various flow-based complex networks has become the key to the analysis of flow-based complex networks. There are a few algorithms for the community detection of the directed graph-based complex network. Most algorithms directly ignore the direction of edges or transform the directed network into the undirected network with weight when detecting communities in the directed complex network. Therefore, they lack effective ability to detect communities in flow-based fraudulent fund transfer networks. Therefore, in this paper, we propose a new concept of the correlation COR between nodes in flow-based fraudulent fund transfer networks, and based on this, a node COR-based local optimization algorithm (LCO) is designed. We adopt our algorithm for the problem of fraudulent funds-raising of the community detection of the victim in flow-based fraudulent fund transfer networks, in which a node represents a transfer record, which has four attributes: payment account, receiving account, transfer time and transfer amount, and node u points to node v with a directed edge, indicating that the receiving account of u is the same as payment account of node v. By analyzing the topological characteristics of each victim’s fund community, our algorithm chooses the node with the least entry degree as the seed node, and expands communities through the correlation between nodes. We verify the effectiveness of the LCO algorithm through the artificially generated directed networks. The results for the artificial directed networks show that our algorithm obtains higher NMI than undirected graph-based overlapping community detection algorithms.","PeriodicalId":236754,"journal":{"name":"2021 International Conference on Computers and Automation (CompAuto)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117173198","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Optimization of Adaptable Prediction and Event- Triggered Replanning using Non-Model Based Methods 基于非模型方法的适应性预测和事件触发重规划优化

2021 International Conference on Computers and Automation (CompAuto) Pub Date : 2021-09-01 DOI: 10.1109/CompAuto54408.2021.00029

Runing Yang, Xilun Zhang, Hengyu Cao, Haoran Peng

引用次数: 0

Solar Radiation Forecasting in Saudi Arabia Using Machine Learning 利用机器学习预测沙特阿拉伯的太阳辐射

2021 International Conference on Computers and Automation (CompAuto) Pub Date : 2021-09-01 DOI: 10.1109/CompAuto54408.2021.00008

Budoor Alwated, T. Brahimi

引用次数: 0

Dynamic model distributing optimization for Mobile Edge Computing 移动边缘计算动态模型分布优化

2021 International Conference on Computers and Automation (CompAuto) Pub Date : 2021-09-01 DOI: 10.1109/CompAuto54408.2021.00012

Changxu Wang, Na Jiang, Libao Chen

引用次数: 0

Food System Model Promotion Research Under Hunger 饥饿条件下粮食系统模型推广研究

2021 International Conference on Computers and Automation (CompAuto) Pub Date : 2021-09-01 DOI: 10.1109/CompAuto54408.2021.00021

Yuyang Chen, Xu Chen, Jiayuan Song, Delong Wu, Chengqi Zhang, Jie Deng

引用次数: 0

Simulating the Key Design Parameters of Oil Field Development 油田开发关键设计参数模拟

2021 International Conference on Computers and Automation (CompAuto) Pub Date : 2021-09-01 DOI: 10.1109/CompAuto54408.2021.00026

M. Tahir, Hongtao Zhou, A. Memon, Wei Liu, Shahidi Ali, Ubedullah Ansari

{"title":"Simulating the Key Design Parameters of Oil Field Development","authors":"M. Tahir, Hongtao Zhou, A. Memon, Wei Liu, Shahidi Ali, Ubedullah Ansari","doi":"10.1109/CompAuto54408.2021.00026","DOIUrl":"https://doi.org/10.1109/CompAuto54408.2021.00026","url":null,"abstract":"The Southwest Betara (SWB) field is located in the western area of the Jabung Block, South Sumatra Basin. The reservoir simulation study was done in this field to predict the reservoir performance, generate production forecast, and finally generate the optimum field of the development plan. To achieve these goals, integrated seismic interpretation, geological model, and reservoir engineering study has been conducted to create dynamic reservoir models. Several reservoir model realizations have been created according to the geological models. The model is validated with the available production history and reservoir pressure data. The PVT data have also been validated with more accurate oil and gas production data. To define optimum field development, sensitivities of the number of producing wells, and pressure maintenance to cumulative oil recovery were done. A total of three cases, base case-11 wells, case-1-18 wells, and case-218 wells plus 2 injectors, were run to optimize the future development of the SWB field. Results revealed that gas production rate in case-1 higher than the other cases. Because of in case-1 the reservoir pressure drops significantly under bubble point pressure. While the cumulative oil production from case-2 is higher than others. On average, the reservoir pressure drops by 800psia after 5 years of production in case-1. In water injection case (case-2), the reservoir pressure drop by 1200psia, it means that water injection can be used to maintain reservoir pressure in the SWB field. An estimated 14.56% oil recovery factor is generated from a reservoir simulation model in base case, 17.8 % from case-1, and 20.69 % from case-2.","PeriodicalId":236754,"journal":{"name":"2021 International Conference on Computers and Automation (CompAuto)","volume":"124 22","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113961641","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Energy usage prediction based on multi-system data for public buildings using machine learning methods 基于机器学习方法的公共建筑多系统数据能源使用预测

2021 International Conference on Computers and Automation (CompAuto) Pub Date : 2021-09-01 DOI: 10.1109/CompAuto54408.2021.00009

Yunlong Li, Yimin Peng, Dengzheng Zhang, Yingan Mai, Zhenrong Ruan, Shufen Liang

引用次数: 0

Preface: CompAuto 2021 前言:CompAuto 2021

2021 International Conference on Computers and Automation (CompAuto) Pub Date : 2021-09-01 DOI: 10.1109/compauto54408.2021.00005

引用次数: 0

Design and Application on Business Data Lineage as a part of Metadata Management 元数据管理中业务数据沿袭的设计与应用

2021 International Conference on Computers and Automation (CompAuto) Pub Date : 2021-09-01 DOI: 10.1109/CompAuto54408.2021.00014

Sona Karkosková, Ota Novotný

引用次数: 0