UAV-Assisted Mobile Edge Computing: Dynamic Trajectory Design and Resource Allocation.

Area of Science:

• * Computer Science
• * Electrical Engineering
• * Robotics

Background:

• * Mobile Edge Computing (MEC) and Unmanned Aerial Vehicles (UAVs) offer flexible computation services beyond terrestrial network coverage.
• * Integrating MEC with UAVs enables enhanced computational support for ground users, particularly in areas with limited infrastructure.
• * Existing systems face challenges in optimizing energy consumption and task completion time.

Purpose of the Study:

• * To develop an efficient algorithm for a UAV-assisted MEC system that minimizes weighted average energy consumption and delay.
• * To jointly optimize user task offloading decisions and UAV trajectory design.
• * To enhance resource management and enable real-time UAV trajectory control.

Main Methods:

• * A novel LQR-Lagrange-based Deep Deterministic Policy Gradient (LLDDPG) algorithm is proposed.
• * The algorithm decomposes the problem into three subproblems: UAV trajectory planning (DDPG), real-time trajectory tracking (LQR), and computational resource allocation (Lagrange duality).
• * Joint optimization of task offloading and UAV trajectory design is achieved through these integrated methods.

Main Results:

• * Simulation results demonstrate the effectiveness of the LLDDPG algorithm.
• * The proposed method significantly improves system resource management in UAV-assisted MEC environments.
• * Real-time UAV trajectory design capabilities are successfully realized.

Conclusions:

• * The LLDDPG algorithm provides an effective solution for optimizing UAV-assisted MEC systems.
• * The approach successfully balances energy consumption and delay, enhancing user experience.
• * This work contributes to the advancement of intelligent resource allocation and trajectory planning in aerial computing systems.