Positioning Autonomous Mobile Robot Based on Measurements Onboard Digital Stereo Vision System

Positioning Autonomous Mobile Robot Based on Measurements Onboard Digital Stereo Vision System

Sergey Valentinovich Kravtsov (Southern Federal University, Rostov-on-Don, Russia) and Konstantin Evgenjevich Rumjantsev (Southern Federal University, Rostov-on-Don, Russia)
Copyright: © 2014 |Pages: 41
DOI: 10.4018/IJRAT.2014070103
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

The problem of local positioning of autonomous mobile robot acting on an unknown scene. The measuring instrument is analyzed on-board stereo vision system consisting of two collinear digital camcorders. The description of the measurement space of digital stereo vision, proposed a stochastic model of the measurement errors of point features scenes. The problem of optimizing the choice of reference for local positioning of autonomous mobile robot. A method for communication dynamics of movement of the mobile robot with the parameters of the digital system stereovision.
Article Preview

1. The Task Of Positioning The Autonomous Mobile Robot And Methods For Its Solution

At the present stage, confident operation of mobile robots can be achieved in a relatively familiar and well-structured workspaces (Kemurdzhian, Gromov, … Kazhukalo, 1993; Moshkin, Petrov, Titov & Yakushenkov, 1990; Collection of scientific articles, 1998). When working in an unfamiliar or changing environment mobile robot should have the ability to adapt to changes in the environment, to respond to unforeseen situations and act on the basis of previous experience. Usually, to solve problems navigation system control robot needs to know the workspace map and the current location of the robot on the map.

Tasks navigation of mobile robots are among the most important problems of governance, where the robot has to do with the changing external environment. The robot must perceive the changing surroundings and to coordinate their actions in accordance with these changes. In this regard, to ensure the accuracy of movement and positioning are important tasks of mobile robot control. The complexity of these tasks is associated with a high level of uncertainty of information about the characteristics of the environment and the mobile robot, noise and errors in the data on-board sensor systems. In all of these problems is the basic problem of autonomous mobile robot location (position) in the workspace.

System to address the problem of positioning a mobile robot can be divided into two classes. This system of global and local positioning.

Global positioning systems allow us to find the location coordinates of the mobile robot in the accepted system of global coordinates. Map of the scene assumed to be known, so navigation is performed in accordance with the purpose of the mission on the map at a known constant update on its position on it. Implementation of global positioning systems requires advanced technical infrastructure and external interfacing with it. Thus the cost of the onboard equipment may be minimized.

By global positioning system should include satellite navigation systems - GPS, GLONASS, Galileo and others; navigation systems for land, sea and air transport infrastructure; positioning system based on cellular networks, and other networks and WiFi.

The disadvantage of global positioning systems is the lack of interaction with the environment. This can lead to undesirable situations unforeseen obstacles when moving around the scene of the inaccuracy (obsolescence) card used, or dynamic changes in the environment.

In addition, global positioning systems have a number of restrictions which significantly affect their use in specific conditions. For example, for satellite navigation systems it is virtually impossible to locate inside buildings, basements and tunnels. The signal strength of satellite systems will be compromised under the cover of foliage of trees and even large clouds. At the reception signal interferes with land-based sources. Positioning system based on cellular networks and WiFi networks are quite low accuracy.

The operating principle of global positioning systems is to use different methods for determining the location, including the well-known method of triangulation, signal-ranging method, or even banal methods of determining binding to the area of responsibility, for example, the base station cellular network or WiFi network.

Local positioning system based on the receipt of information about the environment from on-board sensors. They do not require interaction with the technical infrastructure for autonomous robots are closed for equipment onboard computer complex. Among the advantages of such systems should be noted objectively obtaining information about the environment. According to the on-board sensors built a formalized card, limited-range sensors. Current map coordinate system tied tend to point the position of the mobile robot. Card when moving a mobile robot is constantly refined and updated. The local map can be combined with the global (Mikolajczyk, 2002; Beloglazov, 2010) leased and comparison features scenes, if the pre-known area of operations of the mobile robot. In this case, using a local positioning system can determine the position of a global and effective way navigation plan in accordance with the mission of an autonomous mobile robot.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 5: 2 Issues (2017): 1 Released, 1 Forthcoming
Volume 4: 2 Issues (2016)
Volume 3: 2 Issues (2015)
Volume 2: 2 Issues (2014)
Volume 1: 2 Issues (2013)
View Complete Journal Contents Listing