http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Development of a Human Detecting System Based on Three-Dimensional Point Cloud
Yushi Iwama,Yoshinobu Ando 제어로봇시스템학회 2016 제어로봇시스템학회 국제학술대회 논문집 Vol.2016 No.10
In this paper, we present a human detecting system based on three-dimensional(3D) point cloud by discriminating between human and other objects by using range sensor. Although the shape of same objects seen from a sensor according to the difference of distance between the object and a sensor, we divided some areas in the vicinity of a sensor (about 0.3m~5m) by the difference of distance. Then, we set different detecting conditions in every area. Moreover, we measured shape of human and formulated the detecting condition partly, then we regulated it through experiment in order to detect human regardless of direction of its body. Finally, we tested the system of human detection through some experiments, then we verified its effectiveness even if the height and the place where human is detected are different. This paper details the method we propose and gives some experimental results to verify its effectiveness.
Using GPS outdoor autonomous driving system for robot development(ICCAS2008)
Yutaka Erikawa1,Yoshinobu Ando,Makoto Mizukawa 제어로봇시스템학회 2008 제어로봇시스템학회 국제학술대회 논문집 Vol.2008 No.10
In this paper, we suggest that autonomy moves a robot to the destination without we transcribe a map on interface and order an aim point and the course of the robot on the map. By this, an aim point accords at the coordinate in the map, and it is not necessary to show position information for numerical value. When we produce this system, "The navigation of the robot which it used GPS", "A revision for the difference to occur in GPS and the coordinate value of the map." and "Making of the interface to appoint an aim point from map data." are nominated for necessary element technology. By this report, we performed an element experiment about the navigation of the robot which it used GPS for among element technology.
Jun Konno,Yoshinobu Ando 제어로봇시스템학회 2022 제어로봇시스템학회 국제학술대회 논문집 Vol.2022 No.11
In recent years, robots have been developed that can move autonomously in human environments such as restaurants and airports. For such autonomous mobility, it is important to create a map in advance, and a typical example is Simultaneous Localization and Mapping (SLAM). We created a 3D perception filter that is capable of detecting and eliminating moving point clusters from the input point cloud taken in an indoor environment. In this study, we propose a system that detects moving objects based on camera image recognition and uses the results to construct a more accurate map by minimizing the influence of pedestrians.
Collaboration Technique of RT-Middleware with the Existing Simulator
Tsunehiko FUJITA,Makoto MIZUKAWA,Yoshinobu ANDO 제어로봇시스템학회 2009 제어로봇시스템학회 국제학술대회 논문집 Vol.2009 No.8
The developments of the robots usually depend on their own original specifications. Therefore, the reuse of the robot technologies and/or components is difficult and their development processes required much time and high cost. The National Institute of Advanced Industrial Science and Technology developed RT-Middleware to solve the above problems by modularizing and defining a framework of RT-Component. The development efficiency is improved by using RT-Middleware. However, there is still a problem that there are only few simulators which correspond to RT-Middleware. For the reason, in this paper we focus to simulate the robot, which adopts the RT-Middleware by using an existing simulator. As the result, the system developers can simulate their own system easily.
Improvement of Robustness of the Dual Floor Image Tracking Odometry Measurement
Takahiro TAKASU,Makoto MIZUKAWA,Yoshinobu ANDO 제어로봇시스템학회 2009 제어로봇시스템학회 국제학술대회 논문집 Vol.2009 No.8
Dual Floor Image Tracking (DFIT) is a dead reckoning component, which acquires optical flows of ground images from two cameras. The robot that is equipped with the DFIT can measure its position and rotation without being affected by slip or drift. Noncontact feature of the DFIT enables to add-on robots with an arbitrary driving mechanism. This paper presents the design principle of the DFIT and the features of RT component with experimental evaluation.
Development of Mechatronics Educational Material Applying "Wiki"
Yuya EGUCHI,Yasuhide KUROKAWA,Takaaki OHTA,Yoshinobu ANDO,Makoto MIZUKAWA 제어로봇시스템학회 2008 제어로봇시스템학회 국제학술대회 논문집 Vol.2008 No.10
In this research, we developed and evaluated a web manual of college mechatronics education applying "Wiki". In 2007, we practiced our web manual to the "Microcomputer Seminar" for 3rd grade students who belong to Department of Electrical Engineering, Shibaura Institute of Technology. We obtained the good evaluation for our web manual by means of questionnaires, reports, and presentations.
A Study of Towel Folding by a Robot Arm
Tomoya Oshima,Takashi Yoshimi,Makoto Mizukawa,Yoshinobu Ando 제어로봇시스템학회 2014 제어로봇시스템학회 국제학술대회 논문집 Vol.2014 No.10
In this study, we aim to construct a sequence of towel folding operations by a robot arm. In this paper, we propose the methods for spreading a towel and detecting the grasp positions. For the easy spreading of the towel, we propose how to put the towel on the table. The proposed method lifts one vertex of the towel, and detects the shadow part by using a camera image for recognizing the overlapped condition of the towel. For the folding motion of the towel, we set the vertices of the towel edges as grasping positions. So, we propose the way to detect the vertices of the towel correctly by generating a rectangle which includes all feature points of the towel by using image processing.