پديد آورنده :
مختاري، مريم
عنوان :
بهبود جستجو و پيمايش ربات ها بر پايه شبكه تاريخچه پويش
مقطع تحصيلي :
كارشناسي ارشد
گرايش تحصيلي :
هوش مصنوعي و رباتيك
محل تحصيل :
اصفهان: دانشگاه صنعتي اصفهان ، دانشكده برق و كامپيوتر
صفحه شمار :
سيزده، 174ص: مصور، نمودار، جدول
يادداشت :
ص. ع: به فارسي و انگليسي
استاد راهنما :
مازيار پالهنگ
استاد مشاور :
پژمان خديوي
توصيفگر ها :
جستجو ي هماهنگ , نقشه گراف گرا , طراحي مسير , نقشه سازي , مكان يابي
تاريخ نمايه سازي :
21/11/88
استاد داور :
شادرخ سماوي، رسول موسوي
دانشكده :
مهندسي برق و كامپيوتر
چكيده فارسي :
به فارسي و انگليسي: قابل رويت در نسخه ديجيتال
چكيده انگليسي :
Improving Exploration and Navigation for Multiple Robots Based on Scan History Network Maryam Mokhtari Mokhtari@ec iut ac ir Department of Electrical and Computer Engineering Isfahan University of Technology Isfahan 84156 83111 Iran Degree M Sc Language FarsiSupervisor Dr Maziar Palhang palhang@cc iut ac irAbstract When a disaster like an earthquake occurs human attendance may become dangerousand it is better to use robots for rescue mission A time limitation exists in time of rescuingvictims and robots batteries lives Rescue robots should explore environment as fast aspossible to maximize the chance of finding victims in the limited time Rescue robots sharea map to achieve this goal and to cooperate their exploration The gathered data and itsrepresentation can be an occupancy grid map that shows occupied free or unexploredareas In addition victims locations and the path to reach them can be shown on the map Using multiple robots instead of one robot may lead to some difficulties that should beconsidered like incompatibility in robots information collision of robots to each other andmerging robots maps In this thesis a new map named Scan History Network isintroduced to improve coordination of a team of exploring robots This method uses a moresuitable criterion in selecting exploration targets The benefit of using scan history networkfor coordinating multiple robots over using nearest frontiers as robots goals in occupancygrid map in usual approaches is that robots try to detect less scanned points in theenvironment by keeping a history of robots scans then every robot selects one of thesepoints as its target and tries to plan a suitable path to reach it The target of each robot isselected separated from others with an acceptable distance to have better coverage andexploration of environment Common approaches set more priority for nearer frontierswhen selecting them Although this criterion has been mentioned in the new approach it snot given high priority The main goal of the new approach is to optimize robots exploration and overcoming time limitation and at the same time keeping efficiency Also during map building a Graph Oriented Map in the environment is created in which nodesare junctions in the environment that are the main structural parts in the area This map canbe used in future navigations of robots so repetitive processes are not needed Thereforeexploration and path planning is performed faster An improved approach is introduced todetect junctions In this method unlike usual approaches that obtain graph oriented mapfrom processing occupancy grid map nodes information that are junction locations aregathered directly from the environment by detecting Gaps in laser range scanners data Keywords Coordinated Exploration Multi Robots Scan History Network Graph Oriented Map Path Planning Mapping Localization SLAM
استاد راهنما :
مازيار پالهنگ
استاد مشاور :
پژمان خديوي
استاد داور :
شادرخ سماوي، رسول موسوي
