Enviado por chavabz • 5 de Diciembre de 2013 • Tesis • 1.114 Palabras (5 Páginas) • 169 Visitas
Abstract- Thispaperaimstodescribe the work on multi-robot pattern formation, using vision to recognize them. Robots are assigned to form a specific geometrical shape starting from a random position. A common web camera is used to identify the position and orientation of the robots and using the Open CV libraries and Microsoft Visual C++, all the programming logic is solved to command the robots to move to the desired position. Color filters and figure detectors are used to identify each robot. A group of five physical robots is used to create the desired formations required for this particular project.
Key words: IMECA, ppm, tons, CO, NO2,SO2,O3,PM10,Matlab, Interpolation, Extrapolation.
To apply the knowledge gathered in this course to a real life situation, where only some data is known. Through programming skills use the limited evidence to generate a function able to simulate the general behavior through the years and predict future results, and a program which allows the user to interact with the data.
ulti-robot pattern formation consists of a group of robots that are capable of reaching and maintaining a number of predetermined positions, forming a specific shape as a result.
A multi agent robotic system has many applications. Research projects have incorporated multiple robots for different tasks in different fields, up to and including the entertainment field. The Multi-robot pattern formation can also be very effective in search tasks, mapping tasks, localization tasks, thuscreating more precise and reliable maps.
To develop this project, computational vision is used as the interface between the programming logic and the robots; it is the sensor that determines the actions the robots have to perform to achieve their goal. For computational vision the best definition is: A process that produces, through a group or set of images from the outside world, a description that is useful for the observer and omits all irrelevant information.
From this definition, we have three important aspects we have to consider: 1) Vision is a computer process, 2) the description we obtain depends on the observer and 3) it is necessary to eliminate any non essential information. Image processing is an important field closely linked to that of the computer vision process. Even though both fields have a lot in common, their respective goals are different.
The goal of Image processing is to improve the quality of the image for its later use and interpretation, while the objective of computer vision is to obtain certain features from the image for its computer description and interpretation.
Both vision and image processing are needed to solve the stated problem of the project due to the complex nature of detecting the position and angle of the robots to assign them a task, in this case, reaching a specific position.
Vision and image processing were implemented in the project through the use of Open Cv libraries and Visual C++ software.
OpenCV (Open Computer Vision Library) is an open source library for the purpose of computer vision and image processing. It includes algorithms that permit the application of different tasks, such as face recognition, and identify objects, which is an optimal application for the purposes of this project.
Visual Studio C++ is a software development platform that allows the developer to program using the C++ language. It is useful for creating applications that require a certain interface between the user and the program in question. It is especially