Accuracy-Aware MLLM Task Offloading and Resource Allocation in UAV-Assisted Satellite Edge Computing

Sábháilte in:
Sonraí bibleagrafaíochta
Foilsithe in:Drones vol. 9, no. 7 (2025), p. 500-525
Príomhchruthaitheoir: Yan Huabing
Rannpháirtithe: Huang Hualong, Zhao Zijia, Wang, Zhi, Zhao Zitian
Foilsithe / Cruthaithe:
MDPI AG
Ábhair:
Rochtain ar líne:Citation/Abstract
Full Text + Graphics
Full Text - PDF
Clibeanna: Cuir clib leis
Níl clibeanna ann, Bí ar an gcéad duine le clib a chur leis an taifead seo!

MARC

LEADER 00000nab a2200000uu 4500
001 3233140495
003 UK-CbPIL
022 |a 2504-446X 
024 7 |a 10.3390/drones9070500  |2 doi 
035 |a 3233140495 
045 2 |b d20250101  |b d20251231 
100 1 |a Yan Huabing 
245 1 |a Accuracy-Aware MLLM Task Offloading and Resource Allocation in UAV-Assisted Satellite Edge Computing 
260 |b MDPI AG  |c 2025 
513 |a Journal Article 
520 3 |a This paper presents a novel framework for optimizing multimodal large language model (MLLM) inference through task offloading and resource allocation in UAV-assisted satellite edge computing (SEC) networks. MLLMs leverage transformer architectures to integrate heterogeneous data modalities for IoT applications, particularly real-time monitoring in remote areas. However, cloud computing dependency introduces latency, bandwidth, and privacy challenges, while IoT device limitations require efficient distributed computing solutions. SEC, utilizing low-earth orbit (LEO) satellites and unmanned aerial vehicles (UAVs), extends mobile edge computing to provide ubiquitous computational resources for remote IoTDs. We formulate the joint optimization of MLLM task offloading and resource allocation as a mixed-integer nonlinear programming (MINLP) problem, minimizing latency and energy consumption while optimizing offloading decisions, power allocation, and UAV trajectories. To address the dynamic SEC environment characterized by satellite mobility, we propose an action-decoupled soft actor–critic (AD-SAC) algorithm with discrete–continuous hybrid action spaces. The simulation results demonstrate that our approach significantly outperforms conventional deep reinforcement learning methods in convergence and system cost reduction compared to baseline algorithms. 
653 |a Internet of Things 
653 |a Adaptability 
653 |a Bandwidths 
653 |a Optimization techniques 
653 |a Real time 
653 |a Edge computing 
653 |a Resource allocation 
653 |a Mobile computing 
653 |a Adaptation 
653 |a Remote monitoring 
653 |a Unmanned aerial vehicles 
653 |a Machine learning 
653 |a Low earth orbit satellites 
653 |a Energy consumption 
653 |a Low earth orbits 
653 |a Nonlinear programming 
653 |a Distributed processing 
653 |a Efficiency 
653 |a Scheduling 
653 |a Dynamic programming 
653 |a Large language models 
653 |a Artificial intelligence 
653 |a Cloud computing 
653 |a Decision making 
653 |a Optimization 
653 |a Network latency 
653 |a Computation offloading 
653 |a Variables 
653 |a Algorithms 
653 |a Mixed integer 
700 1 |a Huang Hualong 
700 1 |a Zhao Zijia 
700 1 |a Wang, Zhi 
700 1 |a Zhao Zitian 
773 0 |t Drones  |g vol. 9, no. 7 (2025), p. 500-525 
786 0 |d ProQuest  |t Advanced Technologies & Aerospace Database 
856 4 1 |3 Citation/Abstract  |u https://www.proquest.com/docview/3233140495/abstract/embedded/6A8EOT78XXH2IG52?source=fedsrch 
856 4 0 |3 Full Text + Graphics  |u https://www.proquest.com/docview/3233140495/fulltextwithgraphics/embedded/6A8EOT78XXH2IG52?source=fedsrch 
856 4 0 |3 Full Text - PDF  |u https://www.proquest.com/docview/3233140495/fulltextPDF/embedded/6A8EOT78XXH2IG52?source=fedsrch