2021 International Conference "Nonlinearity, Information and Robotics" (NIR)最新文献_第4页

Real-time static custom gestures recognition based on skeleton hand 基于骨架手的实时静态自定义手势识别

2021 International Conference "Nonlinearity, Information and Robotics" (NIR) Pub Date : 2021-08-26 DOI: 10.1109/NIR52917.2021.9665809

Alexander Osipov, M. Ostanin

引用次数: 7

Increasing performance of image processing algorithms by computing during reception 通过接收过程中的计算提高图像处理算法的性能

2021 International Conference "Nonlinearity, Information and Robotics" (NIR) Pub Date : 2021-08-26 DOI: 10.1109/NIR52917.2021.9666076

Daniil Trihleb, Voronov Roman, V. Artem

引用次数: 0

Tensegrity Topology Merging: Mixed-Integer Convex optimization-based Method 张拉整体拓扑合并:基于混合整数凸优化的方法

2021 International Conference "Nonlinearity, Information and Robotics" (NIR) Pub Date : 2021-08-26 DOI: 10.1109/NIR52917.2021.9666070

R. Khafizov, S. Savin

引用次数: 0

Design and control for the underwater robot with internal rotor 内转子水下机器人的设计与控制

2021 International Conference "Nonlinearity, Information and Robotics" (NIR) Pub Date : 2021-08-26 DOI: 10.1109/NIR52917.2021.9666116

A. Klekovkin, Y. Karavaev, I. Mamaev

引用次数: 0

Development of Fault Accommodation System for Thrusters of Underwater Robots 水下机器人推进器故障调节系统的研制

2021 International Conference "Nonlinearity, Information and Robotics" (NIR) Pub Date : 2021-08-26 DOI: 10.1109/NIR52917.2021.9666119

V. Filaretov, A. Zuev, A. Zhirabok, A. Protsenko

{"title":"Development of Fault Accommodation System for Thrusters of Underwater Robots","authors":"V. Filaretov, A. Zuev, A. Zhirabok, A. Protsenko","doi":"10.1109/NIR52917.2021.9666119","DOIUrl":"https://doi.org/10.1109/NIR52917.2021.9666119","url":null,"abstract":"This paper considers the problem of increasing the reliability of operation of underwater robots through the use of accommodation systems that compensate for the consequences of faults that appear in the thrusters. A new method for constructing accommodation systems, which contains three stages, is proposed. At the first, the detection of emerging faults is carried out using the bank of diagnostic observers. Each observer is synthesized according to a special procedure in such a way that its residual is sensitive to some group of faults and insensitive to others. This makes it possible to isolate each specific fault. At the second stage, the values of sensor errors and deviations of the parameters of the thruster from their nominal values are estimated by additional sliding mode observers. At the third stage, additional control signals for robot’s thrusters are formed. They ensure the stabilization of dynamic properties of thrusters under the faults. The results of simulation are presented, which have confirmed the operability and high efficiency of the synthesized accommodation system.","PeriodicalId":333109,"journal":{"name":"2021 International Conference \"Nonlinearity, Information and Robotics\" (NIR)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129826787","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

Influence of Dissipative Forces and the Design of Mecanum-Wheels on the Omnidirectional Platform Dynamics 耗散力及机轮设计对全向平台动力学的影响

2021 International Conference "Nonlinearity, Information and Robotics" (NIR) Pub Date : 2021-08-26 DOI: 10.1109/NIR52917.2021.9666053

B. Adamov, G. Saypulaev

引用次数: 2

A Study of the Dynamics and Drift of a Micromechanical Gyroscope Taking into Account the Nonlinear Elasticity of the Suspension and Non-orthogonality of the Torsion Axes in the Forced Oscillation Mode 考虑悬架非线性弹性和扭轴非正交性的微机械陀螺仪受迫振荡模式的动力学与漂移研究

2021 International Conference "Nonlinearity, Information and Robotics" (NIR) Pub Date : 2021-08-26 DOI: 10.1109/NIR52917.2021.9666065

M. R. Saypulaev, I. Merkuryev, G. Saypulaev

引用次数: 0

A Classification of Software Defect Prediction Models 软件缺陷预测模型的分类

2021 International Conference "Nonlinearity, Information and Robotics" (NIR) Pub Date : 2021-08-26 DOI: 10.1109/NIR52917.2021.9666110

Sourabh Pal, A. Sillitti

引用次数: 2

Exploration and mapping of an indoor environment using Multirotor Aerial Vehicle 利用多旋翼飞行器探索和测绘室内环境

2021 International Conference "Nonlinearity, Information and Robotics" (NIR) Pub Date : 2021-08-26 DOI: 10.1109/NIR52917.2021.9666129

Mohamed Ahmed Hassan, Geesara Kulathunga, A. Klimchik

{"title":"Exploration and mapping of an indoor environment using Multirotor Aerial Vehicle","authors":"Mohamed Ahmed Hassan, Geesara Kulathunga, A. Klimchik","doi":"10.1109/NIR52917.2021.9666129","DOIUrl":"https://doi.org/10.1109/NIR52917.2021.9666129","url":null,"abstract":"Self-exploration of unknown environments using mobile robots is an important aspect in order to achieve complete automation of the tasks done by robots. The key idea of the exploration is to find a pose at which the robot will gain more information about the environment. By moving from such a pose to another the robot builds a map using the information gained during motion. A frontier based exploration algorithm for 3D spaces is introduced. The algorithm is an extension of the classic frontier exploration algorithm. The space is represented in an octree structure for faster access and efficient memory management. At every iteration of the exploration algorithm, the Multirotor Aerial Vehicle (MAV) performs a 360° rotation to increase the information gain about the environment, this rotation is dependent on the sensor’s field of view. If a rotating sensor which can cover a 360° of the environment is used, the 360° rotation can be skipped without affecting the methodology. The distance moved by the MAV is specified at each iteration of the exploration process depending on the environment. Simulations were conducted to compare the effect of different parameters in the algorithm.","PeriodicalId":333109,"journal":{"name":"2021 International Conference \"Nonlinearity, Information and Robotics\" (NIR)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122823998","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

Continuous learning with random memory for object detection in robotic applications 基于随机记忆的机器人目标检测连续学习

2021 International Conference "Nonlinearity, Information and Robotics" (NIR) Pub Date : 2021-08-26 DOI: 10.1109/NIR52917.2021.9666113

I. Nenakhov, Ruslan Mazhitov, K. Artemov, S. Zabihifar, A. Semochkin, S. Kolyubin

{"title":"Continuous learning with random memory for object detection in robotic applications","authors":"I. Nenakhov, Ruslan Mazhitov, K. Artemov, S. Zabihifar, A. Semochkin, S. Kolyubin","doi":"10.1109/NIR52917.2021.9666113","DOIUrl":"https://doi.org/10.1109/NIR52917.2021.9666113","url":null,"abstract":"If the robot is to interact with environment designed for humans it has to be able to cope with new objects in it’s surrounding, and not only to classify but also effectively localize objects in its reach based on visual sensing. We cannot predict all objects the robot will face. So it needs to learn them while operating, without an external update of software. One of the promising approaches to handle this task is using detectors based on convolution neural networks (CNN) with continuous learning setup. However, there is a well-known problem of catastrophic forgetting (CF) that negatively affects accuracy of neural networks, the main algorithms in object recognition nowadays. Moreover, we need to reach a trade off between training time (because it is unacceptable if the robot trains for hours) and good acccuracy of detection. In this paper we review some of the latest approaches in the field and adapt one of SOtA methods in classification task to detection task. Particularly, we adapt YOLOv5 to work in scenario with continuous learning by adding random memory mechanism from [1] and applying layer freezing in order to increase training speed. We benchmark the proposed method using iCubWorld Transformations dataset and report results with significantly improved metrics and inference speed.","PeriodicalId":333109,"journal":{"name":"2021 International Conference \"Nonlinearity, Information and Robotics\" (NIR)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130247382","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