Please use this identifier to cite or link to this item:
|Title:||Dynamic obstacle detection and avoidance for mobile robots||Authors:||Radhika Ramachandran||Keywords:||DRNTU::Engineering::Electrical and electronic engineering||Issue Date:||2013||Abstract:||Obstacle avoidance is one of the key issues in robot motion planning. It is one of the critical issues especially because of the extensive use of unmanned vehicles in urban areas and for military purposes. During the course of its motion, a robot will have to deal with both static and dynamic obstacles. Some examples of static obstacles are buildings, trees, furniture etc. People, vehicles, other robots etc are the dynamic obstacles that the robot may encounter. For safe and unobstructed motion of the robots, timely detection and avoidance of these obstacles is extremely important. This thesis addresses this issue of obstacle avoidance for a mobile robot equipped with a laser range finder. In this thesis, a real time obstacle avoidance algorithm which can be used for robot navigation in unknown environments has been proposed. The algorithm can be used for avoiding both static and dynamic obstacles. The novelty of this approach lies in the idea of using a polar coordinate environment model for obstacle representation which is combined with a reactive approach to obstacle avoidance to form an efficient and innovative solution to the problem of obstacle avoidance. The range reading from onboard laser range finder is used to find the distance to the obstacle. A window is considered moving along with the robot as it moves. The area within the window is considered to be the region of interest and the obstacles within this area is considered for avoidance. Several experiments were conducted and the algorithm is found to work efficiently and expected results were obtained. The experiment is conducted using a Pioneer 3 robot with Hokuyo laser rangefinder mounted on it. A method to improve the algorithm for fast and non-uniform obstacle motion is also proposed.||URI:||http://hdl.handle.net/10356/55247||Fulltext Permission:||restricted||Fulltext Availability:||With Fulltext|
|Appears in Collections:||EEE Theses|
Page view(s) 10223
checked on Oct 25, 2020
checked on Oct 25, 2020
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.