This academic project constitutes a pivotal element of the ROB-IN initiative, an ambitious endeavor aimed at crafting a robotic companion tailored to assist elderly individuals in their homes. The broader initiative envisions a robot proficient in communication, caregiving, and health monitoring for the elderly. This specific project focuses on endowing a mobile robot with the capability to autonomously locate and grasp household items within dynamic and uncertain environments, leveraging a probabilistic mapping approach and a mobile platform equipped with a dexterous arm. The core objectives revolve around the development and evaluation of various probabilistic strategies, enabling the robot to adeptly search for and grasp household items. The utilization of probabilistic maps plays a central role, in capturing the spatial distribution of objects and key properties such as shape, size, and pose. The project encompasses three integral components: a probabilistic map builder, a map updater, and a map-based planner. These components collectively empower the robot to navigate and manipulate objects intelligently. In addition to the primary focus on probabilistic mapping and object manipulation, the project introduces a novel approach based on the probability of finding the object that we want to find, from the different areas, that divide space around points of interest in space and the probability based on the interrelation between the target object and the proximate entities in those spatial coordinates, according to the relationship that may have the name of the objects. This innovative method seeks to enhance the robot’s decisionmaking capabilities in locating and interacting with household items. The technological foundation incorporates the Stretch RE2 platform, ROS, GENSIM, Pymdptoolbox, and the MS COCO dataset. The interdisciplinary approach blends robotics, probabilistic mapping, and user-centered design, aligning with the broader goals of the ROB-IN initiative. The challenges encountered during ...


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Looking for and grasping household items under probabilistic maps using a mobile platform



    Erscheinungsdatum :

    2024-02-23


    Anmerkungen:

    ETSEIB-240.180456


    Medientyp :

    Hochschulschrift


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Klassifikation :

    DDC:    629




    Looking at Maps

    Peak, K. D. / Royal Institute of Navigation | British Library Conference Proceedings | 1995


    Toward Precise Robotic Grasping by Probabilistic Post-grasp Displacement Estimation

    Zhao, Jialiang Alan / Liang, Jacky / Kroemer, Oliver | TIBKAT | 2021


    GRASPING MECHANISM AND GRASPING DEVICE

    KABASHIMA TAKEFUMI / NISHIDA TAKESHI / SAKAMOTO NAOHITO et al. | Europäisches Patentamt | 2019

    Freier Zugriff

    Stable grasping under pose uncertainty using tactile feedback

    Dang, H. | British Library Online Contents | 2014


    Grasping method by grasping apparatus

    MATSUOKA HIROFUMI | Europäisches Patentamt | 2016

    Freier Zugriff