Vides autonomas apzināšanas simulācija UGV un UAV sistēmai
Loading...
Date
Authors
Advisor
Journal Title
Journal ISSN
Volume Title
Publisher
Latvijas Universitāte
Language
lav
Abstract
Vides autonomās apzināšana ir plaši izplatīta tēma robotikas nozarē. Šajā darbā uzsvars tiek likts uz vides autonomo apzināšanu, izmantojot heterogēnu robotu sistēmu, konkrētāk, UGV un UAV sistēmu. Vides autonomā apzināšana izpaužas 3D vides kartēšanā, 2D lokālo režģkaršu iegūšanā, iegūto režģkaršu apvienošanā un maršrutu punktu iegūšanā no apvienotās globālās režģkartes. Maršrutu punktu iegūšanai tiek izmantotas dziļās stimulētās mācīšanās metodes.
Autonomous exploration is a popular topic in robotics field. In this work, the focus is on the autonomous exploration, using a heterogenous robot system, specifically a system consisting of a UGV and a UAV. Autonomous exploration is done by mapping the 3D environment, creating 2D local occupancy grid maps, and merging them together into a global occupancy grid map. From the global map, the waypoints for the robot navigation are calculated. Waypoint calculation is done by using deep reinforcement learning.
Autonomous exploration is a popular topic in robotics field. In this work, the focus is on the autonomous exploration, using a heterogenous robot system, specifically a system consisting of a UGV and a UAV. Autonomous exploration is done by mapping the 3D environment, creating 2D local occupancy grid maps, and merging them together into a global occupancy grid map. From the global map, the waypoints for the robot navigation are calculated. Waypoint calculation is done by using deep reinforcement learning.