Hierarchical Task Offloading for UAV-Assisted Vehicular Edge Computing via Deep Reinforcement Learning

TL;DR

This paper introduces a hierarchical deep reinforcement learning approach for UAV-assisted vehicular edge computing, optimizing task offloading, trajectory, and resource coordination to improve delay, energy efficiency, and robustness in dynamic networks.

Contribution

It proposes a novel dual-layer UAV-assisted architecture with a hierarchical offloading scheme using deep reinforcement learning to coordinate heterogeneous resources and adapt to network dynamics.

Findings

Outperforms baseline methods in task completion rate

Enhances system efficiency and convergence speed

Demonstrates robustness in dynamic vehicular environments

Abstract

With the emergence of compute-intensive and delay-sensitive applications in vehicular networks, unmanned aerial vehicles (UAVs) have emerged as a promising complement for vehicular edge computing due to the high mobility and flexible deployment. However, the existing UAV-assisted offloading strategies are insufficient in coordinating heterogeneous computing resources and adapting to dynamic network conditions. Hence, this paper proposes a dual-layer UAV-assisted edge computing architecture based on partial offloading, composed of the relay capability of high-altitude UAVs and the computing support of low-altitude UAVs. The proposed architecture enables efficient integration and coordination of heterogeneous resources. A joint optimization problem is formulated to minimize the system delay and energy consumption while ensuring the task completion rate. To solve the high-dimensional decision problem, we reformulate the problem as a Markov decision process and propose a hierarchical offloading scheme based on the soft actor-critic algorithm. The method decouples global and local decisions, where the global decisions integrate offloading ratios and trajectory planning into continuous actions, while the local scheduling is handled via designing a priority-based mechanism. Simulations are conducted and demonstrate that the proposed approach outperforms several baselines in task completion rate, system efficiency, and convergence speed, showing strong robustness and applicability in dynamic vehicular environments.