发布求助
文献互助 智能选刊 最新文献

2021 25th International Conference on Methods and Models in Automation and Robotics (MMAR)最新文献

筛选
英文 中文
Weight Perturbation as a Method for Improving Performance of Deep Neural Networks 权扰动作为一种提高深度神经网络性能的方法
2021 25th International Conference on Methods and Models in Automation and Robotics (MMAR) Pub Date : 2021-08-23 DOI: 10.1109/MMAR49549.2021.9528460
M. Nowak, Kamil Lelowicz
{"title":"Weight Perturbation as a Method for Improving Performance of Deep Neural Networks","authors":"M. Nowak, Kamil Lelowicz","doi":"10.1109/MMAR49549.2021.9528460","DOIUrl":"https://doi.org/10.1109/MMAR49549.2021.9528460","url":null,"abstract":"Learnable weights in very deep neural networks are not utilized in an efficient way i.e. the variance of weights in some non-performing layers is close to zero. One approach taken to mitigate this problem, has been the connection pruning, in which some of the weights in a network are masked with zeros, without negatively affecting the network performance. The drawback of such approach is that in the GPU implementations of neural networks, pruning the weights does not affect the network training or inference speed. In this paper we propose an alternative solution to the problem of inefficient utilization of network learning capability - perturbing the weights of the network during training with additive, zero-mean noise. It causes the variance of weights in non-performing layers to rise, and can improve the performance of neural networks. In this paper we show the effect of the weight perturbation on the performance of five well known network architectures, while testing them on the task of CIFAR100 classification.","PeriodicalId":212910,"journal":{"name":"2021 25th International Conference on Methods and Models in Automation and Robotics (MMAR)","volume":"127 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131560192","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
On Assignment of the Upper Bohl Exponent for Linear Discrete-Time Systems in Infinite-Dimensional Spaces 无限维空间中线性离散系统上Bohl指数的赋值问题
2021 25th International Conference on Methods and Models in Automation and Robotics (MMAR) Pub Date : 2021-08-23 DOI: 10.1109/MMAR49549.2021.9528496
E. Makarov, M. Niezabitowski, S. Popova, V. Zaitsev, Marina Zhuravleva
{"title":"On Assignment of the Upper Bohl Exponent for Linear Discrete-Time Systems in Infinite-Dimensional Spaces","authors":"E. Makarov, M. Niezabitowski, S. Popova, V. Zaitsev, Marina Zhuravleva","doi":"10.1109/MMAR49549.2021.9528496","DOIUrl":"https://doi.org/10.1109/MMAR49549.2021.9528496","url":null,"abstract":"In this paper we study linear discrete-time control systems in infinite-dimensional spaces with periodically time-varying bounded input and output operators. For such systems we consider a problem of arbitrary assignability of the upper Bohl exponent by non-stationary linear static state feedback and by non-stationary linear dynamic output feedback. This is a generalization of the stabilization problem, which has been well studied for systems of various types. The main results present sufficient conditions for considered types of assignability.","PeriodicalId":212910,"journal":{"name":"2021 25th International Conference on Methods and Models in Automation and Robotics (MMAR)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131578322","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: 2
Virtual Urban Space Simulator — — Gliwice, Poland City Centre Example in the Context of COVID-19 pandemic 虚拟城市空间模拟器——2019冠状病毒病大流行背景下波兰市中心格列维采的例子
2021 25th International Conference on Methods and Models in Automation and Robotics (MMAR) Pub Date : 2021-08-23 DOI: 10.1109/MMAR49549.2021.9528467
A. Gałuszka, Eryka Probierz, D. Bereska, Marcin Pacholczyk, Ewa Bajer, J. Śmieja, Kamil Skowronski, Aleksandra Szydłowka, P. Solarz, Lukasz Kania, M. Cholewa, Pawel Woronow
{"title":"Virtual Urban Space Simulator — — Gliwice, Poland City Centre Example in the Context of COVID-19 pandemic","authors":"A. Gałuszka, Eryka Probierz, D. Bereska, Marcin Pacholczyk, Ewa Bajer, J. Śmieja, Kamil Skowronski, Aleksandra Szydłowka, P. Solarz, Lukasz Kania, M. Cholewa, Pawel Woronow","doi":"10.1109/MMAR49549.2021.9528467","DOIUrl":"https://doi.org/10.1109/MMAR49549.2021.9528467","url":null,"abstract":"Virtual environments (VE) are commonly used in learning or training in many different areas. We are focused on the problem when a group of artificial agents are simulated at the same time; their behavior is easily recognized as ‘artificial’. To avoid this undesired property presented work was aimed at creating a virtual world with agents equipped with psychosomatic elements that occur in the real world. It has been chosen the implementation of basic needs based on Maslow's hierarchy of needs (MHN), since they are one of necessary elements for the development of simulation of the real world. It is shown that the VE design based on artificial intelligence (AI) planning algorithms and the theory of MHN can be efficiently applied to generate semi-realistic behavior of agents population. Models of coronavirus spread have been introduced into the simulated environment. It allowed to build cases related to the number of infected people and the rate of infection depending on the level of avatar activity associated with basic needs.","PeriodicalId":212910,"journal":{"name":"2021 25th International Conference on Methods and Models in Automation and Robotics (MMAR)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127134965","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
Virtual simulation environment of an autonomous A-EVE vehicle 自动驾驶车辆虚拟仿真环境
2021 25th International Conference on Methods and Models in Automation and Robotics (MMAR) Pub Date : 2021-08-23 DOI: 10.1109/MMAR49549.2021.9528440
M. Długosz, M. Roman, P. Węgrzyn
{"title":"Virtual simulation environment of an autonomous A-EVE vehicle","authors":"M. Długosz, M. Roman, P. Węgrzyn","doi":"10.1109/MMAR49549.2021.9528440","DOIUrl":"https://doi.org/10.1109/MMAR49549.2021.9528440","url":null,"abstract":"The base condition which has to be met by autonomous cars is the guaranteed safety of passengers and other participants. Therefore, it is necessary to intensively test devices and algorithms for autonomous driving, what generates high costs. In order to reduce costs, it is necessary to perform as many tests as possible in a virtual environment with the use of simulators. The article describes the virtual simulation environment of an experimental autonomous vehicle called A-EVE. The GAZEBO simulator was used to implement the dynamic and kinematic model and sensors of the A-EVE vehicle. The use of the virtual GAZEBO environment enables rapid prototyping and testing of new functionalities of the A-EVE vehicle and the reduction of the number of practical experiments to the necessary minimum.","PeriodicalId":212910,"journal":{"name":"2021 25th International Conference on Methods and Models in Automation and Robotics (MMAR)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129669281","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
The Use of Counting Peaks Method for the Purpose of Smoothing Filtering Efficiency Assessment in Analysis of Electroencephalography Data 计数峰法在脑电图数据分析中平滑滤波效率评价的应用
2021 25th International Conference on Methods and Models in Automation and Robotics (MMAR) Pub Date : 2021-08-23 DOI: 10.1109/MMAR49549.2021.9528429
Aleksandra Kawala-Sterniuk, Natalia Browarska, J. Zygarlicki, M. Kaminski, Barbara Grochowicz, Mariusz Pelc
{"title":"The Use of Counting Peaks Method for the Purpose of Smoothing Filtering Efficiency Assessment in Analysis of Electroencephalography Data","authors":"Aleksandra Kawala-Sterniuk, Natalia Browarska, J. Zygarlicki, M. Kaminski, Barbara Grochowicz, Mariusz Pelc","doi":"10.1109/MMAR49549.2021.9528429","DOIUrl":"https://doi.org/10.1109/MMAR49549.2021.9528429","url":null,"abstract":"In this paper an innovative counting peaks method applied for the purpose of efficiency assessment of smoothing filtering applied on electroencephalography data was presented. The proposed method gave promising results and can be treated as an introductory method for intelligent machine-based reasoning. The analysed data were obtained from the open source data-base and recorded with the OpenBCI EEG headset. The comparison focused on Moving Average and Savitzky-Golay (SG) filter. The best results were obtained with the use of the Moving Average filtering.","PeriodicalId":212910,"journal":{"name":"2021 25th International Conference on Methods and Models in Automation and Robotics (MMAR)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120960374","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: 2
MMAR 2021 Author Index mmar2021作者索引
2021 25th International Conference on Methods and Models in Automation and Robotics (MMAR) Pub Date : 2021-08-23 DOI: 10.1109/mmar49549.2021.9528453
{"title":"MMAR 2021 Author Index","authors":"","doi":"10.1109/mmar49549.2021.9528453","DOIUrl":"https://doi.org/10.1109/mmar49549.2021.9528453","url":null,"abstract":"","PeriodicalId":212910,"journal":{"name":"2021 25th International Conference on Methods and Models in Automation and Robotics (MMAR)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126005205","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
A simple vision-based navigation and control strategy for autonomous drone racing 一个简单的基于视觉的自主无人机比赛导航和控制策略
2021 25th International Conference on Methods and Models in Automation and Robotics (MMAR) Pub Date : 2021-04-20 DOI: 10.36227/TECHRXIV.14452725.V1
Artur Cyba, Hubert Szolc, T. Kryjak
{"title":"A simple vision-based navigation and control strategy for autonomous drone racing","authors":"Artur Cyba, Hubert Szolc, T. Kryjak","doi":"10.36227/TECHRXIV.14452725.V1","DOIUrl":"https://doi.org/10.36227/TECHRXIV.14452725.V1","url":null,"abstract":"In this paper, we present a control system that allows a drone to fly autonomously through a series of gates marked with ArUco tags. A simple and low-cost DJI Tello EDU quad-rotor platform was used. Based on the API provided by the manufacturer, we have created a Python application that enables the communication with the drone over WiFi, realises drone positioning based on visual feedback, and generates control. Two control strategies were proposed, compared and critically analysed. In addition, the accuracy of the positioning method used was measured. The application was evaluated on a laptop computer (about 40 fps) and a Nvidia Jetson TX2 embedded GPU platform (about 25 fps). We provide the developed code on GitHub.","PeriodicalId":212910,"journal":{"name":"2021 25th International Conference on Methods and Models in Automation and Robotics (MMAR)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116626679","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: 3
Transmitter Discovery through Radio-Visual Probabilistic Active Sensing 基于无线电视觉概率主动感知的发射机发现
2021 25th International Conference on Methods and Models in Automation and Robotics (MMAR) Pub Date : 2021-03-27 DOI: 10.1109/MMAR49549.2021.9528477
Luca Varotto, A. Cenedese
{"title":"Transmitter Discovery through Radio-Visual Probabilistic Active Sensing","authors":"Luca Varotto, A. Cenedese","doi":"10.1109/MMAR49549.2021.9528477","DOIUrl":"https://doi.org/10.1109/MMAR49549.2021.9528477","url":null,"abstract":"Multi-modal Probabilistic Active Sensing (MMPAS) uses sensor fusion and probabilistic models to control the perception process of robotic sensing platforms. MMPAS is successfully employed in environmental exploration, collaborative mobile robotics, and target tracking, being fostered by the high performance guarantees on autonomous perception. In this context, we propose a bi-Radio-Visual PAS scheme (Ra2ViPAS) to solve the transmitter discovery problem. Specifically, we firstly exploit the correlation between radio and visual measurements to learn a target detection model in a self-supervised manner. Then, the model is combined with antenna radiation anisotropies into a Bayesian Optimization framework that controls the platform. We show that the proposed algorithm attains an accuracy of 92%, overcoming two other probabilistic active sensing baselines.","PeriodicalId":212910,"journal":{"name":"2021 25th International Conference on Methods and Models in Automation and Robotics (MMAR)","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127357086","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: 2
首页 上一页
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整，为了账户安全，请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们：info@booksci.cn Book学术提供免费学术资源搜索服务，方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号：604180095
Book学术官方微信
小红书