Abstract
dc:descriptionNavigation and collaborative mapping of a team of mobile robots is the subject of this dissertation. Navigation is a basic component that needs to be dealt with for the mobile robots in order to execute autonomously the multitude of the tasks that are assigned to. Given an area of interest, an initial position and a final target position for a robot, a path must be found that satisfies the kinematic constraints initially imposed by the environment, and then the dynamic constraints due to the robot model and the constraints in the system input. Initially, in the dissertation, the navigation in known environments was studied. The problem was decomposed in two parts, finding the path initially and then the extraction of the trajectory from it. For the path planning, artificial potential fields (in the case of the planar environment) and navigation functions (in the case of a 2D surface) were used. For the trajectory extraction, the method was based upon Bèzier curves. In the sequel, the focus was shifted towards unknown planar environments. In this case, the robot has a range sensor for the construction for the area map. Given the mapping, a switching navigation function is developed, where initially the exploration of the area towards the discovery of the target position is performed, relying on frontier based exploration. If the target is discovered, then the navigation function switches to guide it towards it. Initially a limited ranged circular sensing pattern was assumed, with a simple kinematic model for the robot and perfect knowledge of the robot position. Afterwards this method was improved for a circular sector sensing pattern, a more complete robot model, and the introduction of the pose uncertainty. Lastly the transition to a team of mobile robots occurred. By preserving the basic principle of the method that was developed for unknown environments, the collaborative mapping and creation of a global map from the team was introduced as a concept in order to improve the motion towards the individual target that was assigned for each member, with the transition to a switching frontier cost function.
Degree
thesis:*- Grantor dc:publisher
- University of Patras
- Year dc:date
- 2017
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Arvanitakis, Ioannis
- Αρβανιτάκης, Ιωάννης
Subjects
dc:subject × 20- Πλοήγηση
- Κινούμενα ρομπότ
- Αποφυγή εμποδίων
- Σχεδιασμός κίνησης
- Συνάρτηση πλοήγησης
- Εξερεύνηση
- Χαρτογράφηση
- Ρομποτικό σμήνος
- Navigation
- Mobile robots
- Obstacle avoidance
- Motion planning
- Navigation functions
- Exploration
- Mapping
- Swarm robotics
- 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/41183