Bronnikov A. Models and methods of robots adaptive visual control

Українська версія

Thesis for the degree of Candidate of Sciences (CSc)

State registration number

0421U103532

Applicant for

Specialization

  • 05.13.07 - Автоматизація процесів керування

23-09-2021

Specialized Academic Board

Д 64.052.08

Kharkiv National University Of Radio Electronics

Essay

The object is control processes of flexible integrated robotic radio-electronic manufacturing. The goal – development of new and improvement of existing models and methods of adaptive visual control, providing an increase in the productivity, speed, and reliability of robotic means of flexible integrated production systems. Methods – methods of set theory, methods of automatic control theory and fuzzy sets, methods of image processing theory. Results – functional FIS model, which provides a description of objects interaction at the level of production performance, transport, service, monitoring, control and other tasks assigned to FIS, and, unlike existing ones, is based on a declarative model of the workspace takes into account the features of technological operations, contains a detailed description of objects properties and their interactions, including the possibility of quantifying the cost and quality of work was firstly developed; a model for controlling a mobile robot in the FIS space, controls the chassis of a two-wheeled operation, and, unlike existing ones, operates on the basis of information from object / local computer vision systems and performs the functions of FIS objects recognition and identification, provides the calculation of coordinates and speeds of a robotic platform and other FIS objects, calculation of linear and curvilinear movements trajectories in a discrete workspace, presentation of the route in the movements sequences form, ensures the organization of transport operations control by their visual position was improved; the method of adaptive visual control has been improved, it describes a set of techniques and operations that should be applied in intelligent control systems of robots for planning a movement task based on obtaining visual information from computer (technical) vision systems, and, unlike existing methods, is focused on using mobile robotic systems; the model for controlling a mobile robot in the FIS space was further developed, it controls the chassis of a two-wheeled operation, and, unlike existing ones, operates on the basis of information from global / local computer vision systems and performs the functions of recognition and identification of FIS objects, provides the calculation of coordinates and velocities a robotic platform and other FIS objects, calculation of linear and curvilinear trajectories of movements in a discrete workspace, presentation of a route in the form of sequences of movements, ensures the organization of control of transporting works according to their visual position; the fuzzy component of the model contains a detailed description of the procedures for the formation of linguistic variables associated with the dynamics of the movement of the mobile robotic platform and formalizes the set of products of the fuzzy controller.

Files

Similar theses