بهبود الگوريتم مكان يابي و رسم نقشه به صورت همزمان با استفاده از صافي كالمن توسعه يافته براي يك روبات پايه متحرك

چكيده انگليسي :

994 Improving simultaneously localization and mapping algorithm with extended Kalman filter for a mobile robot Yasin Zamani y zamani@ec iut ac ir Date of Submission 2012 9 17 Department of Electrical and Computer Engineering Isfahan University of Technology Isfahan 84156 83111 Iran Degree M Sc Language FarsiSupervisor Mohammad Ali Montazeri montazeri@cc iut ac irAbstractRobots generally divided into two categories based mobile and fixed base Movable base or mobilerobots cover a wide range of human needs Mobile robots are used for planetary exploration exploring areas that are harmful to humans and some other applications One of the main goals inthe field of autonomous mobile robots is proposed in order to explore an unknown environment robot generated map of the environment detect the presence of specific symptoms in environment vital life signs warning signs and risk factors environmental pollution and etc and accurate mapof their location in the environment To achieve this goal the fundamental issues are determineunexplored areas of the environment routing to guide the robot to identify passable routes fordelivering the robot to unexplored areas and avoid of obstacles and robot control Scientific issuesrelated to mobile robotics includes subjects such as their understanding of the environment andhow show it Traveling and exploring the unknown environment and changing environment withartificial agents which are controlled by the computers There are uncertainties in solving problemsin mobile robotics Generally this phenomenon is permanently present in many robotic problems in some cases can be irrespective and in the other issues to deal with them with the appropriatetools A branch of the robotics which has been considered in the recent decades is a probabilisticrobotic This field of robotic explores issues which involved with such uncertainties Simultaneously localization and mapping raised this question that is it possible for an autonomousvehicle starting to move in an unknown location in an unknown environment well and thengradually begin to build maps of the surrounding environment while both of these maps are alsoused to calculate your exact location The most common approach for solving SLAM is usingtechniques based on probability theory and Kalman filter In the past decade one of the issuesabout mobile robots that researchers have been interested in is probabilistic robotic Thisknowledge involved in the robotics issues that the uncertainty exists in them Simultaneouslylocalization and mapping is one of the most important topics of probabilistic robotics and isessential for mobile robots In this thesis the simultaneously localization and mapping algorithmusing extended Kalman filter EKF SLAM is implemented on a rescue mobile robot Eliminateconfusion within the data collection which had been in the accuracy range of the laser range finder LRF sensor of the robots and using an angular threshold was improved the line extractionalgorithm of the mapping Also to improve the performance of iterative closest point ICP algorithm in the association step instead of the nearest neighbor search NNS method we usedthe methods for solving the assignment problem Another innovation of this thesis is to improveestimated transformation by rotary encoder sensor before its use in initial prediction step of EKF SLAM by ICP algorithm This improved process is called iEKF SLAM algorithm It is possibleto show that the two dimensional concept localization and mapping extend to three dimensions Forthis purpose data extracted from 3D laser scanner must be used Keywords Mobile robots Laser range finder Localization Mapping

زماني، ياسين

بهبود الگوريتم مكان يابي و رسم نقشه به صورت همزمان با استفاده از صافي كالمن توسعه يافته براي يك روبات پايه متحرك

پويشگر ليزري , رسم نقشه