Task Offloading Algorithm for Multiple Unmanned Aerial Vehicles Based on Temporal Graph.

Area of Science:

• Computer Science
• Electrical Engineering
• Network Engineering

Background:

• The increasing scale of data necessitates efficient task offloading in future 6G networks.
• Unmanned Aerial Vehicles (UAVs) offer potential for mobile edge computing but face limitations in battery and computation.
• Collaborative UAVs can overcome individual limitations for enhanced task offloading.

Purpose of the Study:

• To propose a novel task offloading algorithm for multiple UAVs using a temporal graph.
• To minimize the total completion time for UAV swarm task offloading.
• To optimize task prioritization, subtask dependencies, and resource allocation.

Main Methods:

• Formulated an optimization problem to minimize total completion time.
• Introduced a temporal graph to model service nodes and task sequences.
• Calculated proximity indices and timestamp offsets to determine task priorities.
• Transformed the problem into a directed acyclic graph connectivity problem.
• Proposed a two-stage matching algorithm for optimal resource allocation.

Main Results:

• The proposed algorithm significantly reduces task completion times for UAV swarms.
• Simulation results show superior performance compared to existing algorithms.
• The algorithm effectively manages task utility and resource allocation.

Conclusions:

• The temporal graph-based algorithm provides an effective solution for UAV task offloading in 6G networks.
• Optimized task offloading enhances the efficiency and utility of UAV swarms.
• The approach addresses computational and battery limitations of individual UAVs.