National Technical University of Athens (NTUA)
Μεθοδολογία ανάπτυξης αντικειμενοστραφούς λογσιμικού για ρομποτικά συστήματα με ανατροφοφότηση από αισθητήρες
Abstract
dc:descriptionThis work presents a new methodology for the development of software that allows the integration of external sensors in open robotic systems that are controlled at the end- effector level. The methodology defines a process that leads to the creation of a library of modular and reusable objects, which are intended for composition of robotic systems. By implementing new reusable objects for the needs of new applications the library becomes enriched. The development process follows the iterative model of requirements-analysis- design-implementation-testing. The methodology, which is based on an open architecture, enables easy and fast development. The modularity and re-use that is achieved through the object oriented approach enable the production of high quality software at the control objects level. The proposed architecture includes the subsystems of Sensor Management SM, State Error Calculator SEC, Sensor - Actuator Interface SAI and Actuator Management AM. The AM determines the next system state preventing under proper conditions the oscillation of the end-effector. The coupling with the sub systems that operate based on the sensor cycle is done through the SAI subsystem. The SM sub system gives the possibility of integration of the most popular sensors, (proximity, force-torque, cameras, tactile and range sensors). The SEC subsystem extracts the state error at the end effector level, using the measurements that are received from SM. The methodology provides significant benefits to applications that employ external sensors. It achieves the decoupling of the subsystems design from their use and provides the principles of modular implementation of particular applications that are expressed by the proposed open architecture. The library objects are reusable and thus provide benefits such as speed of development, high quality and low cost. The library is extensible quantitatively, by adding new objects or sub systems, as well as qualitatively, by incorporating improved algorithms. Therefore the new control and pattern recognition algorithms can be quickly and reliably incorporated and a closer connection between research and industry can be established. The methodology was verified by composing industrial applications that can be used as a paradigm for the implementation of new systems. The first of them is related to the automatic sunroof fitting onto car bodies with use of feedback from 4 cameras, while the second application executes automated measurement of gaps on car shells by employing stereo vision. The third application concerns a robot for visual inspection of gaps on vehicles and can be implemented through reuse of sub systems. The experimental results show that with proper parameterization the applications operate according to the requirements, thus proving the validity of the methodology.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2002
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Kosmopoulos, Dimitrios
- Κοσμόπουλος, Δημήτριος
Subjects
dc:subject × 22- Ανατροφοδότηση από εξωτερικούς αισθητήρες
- Έλεγχος κλειστού βρόχου στο επίπεδο τελικού στοιχείου δράσης
- Ανοικτή αρχιτεκτονική ρομπτοτικού ελεγκτή
- Βιομηχανικά ρομπότ
- Βιβλιοθήκη αντικειμένων ελέγχου
- Επαναχρησιμοποίηση λογισμικού
- Μεθοδολογία ανάπτυξης λογισμικού
- Μέτρηση διακένων
- Προσαρμογή ηλιοροφής
- External sensory feedback
- Closed loop control at the end-effector level
- Open robot controller architecture
- Industrial robots
- Library of control objects
- Software reuse
- Software development methodology
- Gap measurement
- Sunroof fitting
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Engineering and Technology
- Electrical Engineering, Electronic Engineering, Information Engineering
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/16965