Graduate School and Research Center in Digital Sciences

UAV path planning for wireless data harvesting: A deep reinforcement learning approach

Bayerlein, Harald; Theile, Mirco; Caccamo, Marco; Gesbert, David

Submitted to GLOBECOM 2020, IEEE Global Communications Conference, 7-11 December 2020, Taipei, Taiwan

Autonomous deployment of unmanned aerial vehicles (UAVs) supporting next-generation communication networks requires efficient trajectory planning methods. We propose a new end-to-end reinforcement learning (RL) approach to UAV-enabled data collection from Internet of Things (IoT) devices in an urban environment. An autonomous drone is tasked with gathering data from distributed sensor nodes subject to limited flying time and obstacle avoidance. While previous approaches, learning and non-learning based, must perform expensive recomputations or relearn a behavior when important scenario parameters such as the number of sensors, sensor positions, or maximum flying time, change, we train a double deep Q-network (DDQN) with combined experience replay to learn a UAV control policy that generalizes over changing scenario parameters. By exploiting a multi-layer map of the environment fed through convolutional network layers to the agent, we show that our proposed network architecture enables the agent to make movement decisions for a variety of scenario parameters that balance the data collection goal with flight time efficiency and safety constraints. Considerable advantages in learning efficiency from using a map centered on the UAV's position over a non-centered map are also illustrated.    

Arxiv Bibtex

Title:UAV path planning for wireless data harvesting: A deep reinforcement learning approach
Type:Conference
Language:English
City:Taipei
Country:TAIWAN, PROVINCE OF CHINA
Date:
Department:Communication systems
Eurecom ref:6302
Copyright: © EURECOM. Personal use of this material is permitted. The definitive version of this paper was published in Submitted to GLOBECOM 2020, IEEE Global Communications Conference, 7-11 December 2020, Taipei, Taiwan and is available at :
Bibtex: @inproceedings{EURECOM+6302, year = {2020}, title = {{UAV} path planning for wireless data harvesting: {A} deep reinforcement learning approach}, author = {{B}ayerlein, {H}arald and {T}heile, {M}irco and {C}accamo, {M}arco and {G}esbert, {D}avid}, booktitle = {{S}ubmitted to {GLOBECOM} 2020, {IEEE} {G}lobal {C}ommunications {C}onference, 7-11 {D}ecember 2020, {T}aipei, {T}aiwan}, address = {{T}aipei, {TAIWAN}, {PROVINCE} {OF} {CHINA}}, month = {12}, url = {http://www.eurecom.fr/publication/6302} }
See also: