Democritus University of Thrace (DUTH)
Cognitive human-robot interaction with social maps
Abstract
dc:descriptionWith the robots tending to accumulate more and more capabilities beyond thelevel of acting in a deterministic fashion, the idea of introducing them intoour every day lives seems to be closer now. Robotics systems and techniquesappeared during the recent years have achieved astonishing potential to perceiveand interpret their surrounding not only as low level features but alsoclose to human understandable concepts. Such advances, in conjunction withthe aspiration to incorporate robots into domestic or public places, led to theflourishing of fields dealing with their response in human presence. Followingthis notion, the field of social mapping was recently introduced in order tomanage the shared space among robots and individuals in an ordinary fashion.Social mapping deals with the problem of human aware robot navigation. Whilethe scope of autonomous navigation is constrained to obstacle avoidance andreaching the goal, the social one aims to navigate by additionally consideringfactors such as human comfort, naturalness and sociability. Such maps operatejointly with the metric ones, thus endowing the robot’s perception aboutthe environment particularly about individuals, their relative placement withrespect to the robot as well as among themselves in the map and provide theopportunity for the prosperity of robot systems that administer the sense offamiliarity towards the users, leading to the broad acceptance of robots in everyday lives. The main scope of this PhD dissertation is the development of preciseand persistent social maps which will allow the operation of robot in humanpopulated environments.
Degree
thesis:*- Grantor dc:publisher
- Democritus University of Thrace (DUTH)
- Year dc:date
- 2016
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Charalampous, Konstantinos
- Χαραλάμπους, Κωνσταντίνος
Subjects
dc:subject × 6Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/38611