Graduate School and Research Center in Digital Sciences

Learning to rest: A Q-learning approach to flying base station trajectory design with landing spots

Bayerlein, Harald

Invited talk in the Doctoral Seminar on Methods of Signal Processing, Technical University of Munich, 21 March 2019, Munich, Germany

This talk considers the problem of trajectory optimization for an autonomous UAV-mounted base station that provides communication services to ground users with the aid of landing spots (LSs). Recently, the concept of LSs was introduced to alleviate the problem of short mission durations arising from the limited on-board battery budget of the UAV, which severely limits network performance. In this work, using Q-learning, a model-free reinforcement learning (RL) technique, a neural network (NN) was trained to make movement decisions for the UAV that maximize the data collected from the ground users while minimizing power consumption by exploiting the landing spots. The system is thus able to intelligently integrate landing spots into the trajectory to extend flying time without any explicit information about the environment.

Bibtex

Title:Learning to rest: A Q-learning approach to flying base station trajectory design with landing spots
Type:Talk
Language:English
City:Munich
Country:GERMANY
Date:
Department:Communication systems
Eurecom ref:5843
Copyright: © EURECOM. Personal use of this material is permitted. The definitive version of this paper was published in Invited talk in the Doctoral Seminar on Methods of Signal Processing, Technical University of Munich, 21 March 2019, Munich, Germany and is available at :
Bibtex: @talk{EURECOM+5843, year = {2019}, title = {{L}earning to rest: {A} {Q}-learning approach to flying base station trajectory design with landing spots}, author = {{B}ayerlein, {H}arald}, number = {EURECOM+5843}, month = {03}, institution = {Eurecom} address = {{M}unich, {GERMANY}}, url = {http://www.eurecom.fr/publication/5843} }
See also: