SAC-Based Offloading and Resource Allocation Optimization for a Low-Altitude Economy Network

Driven by the prosperous vision of 6G-empowered low-altitude economy (LAE) network, integration with mobile edge computing (MEC) is expected to primarily address the trade-off between minimising the task delay and the energy consumption of the system network. This paper proposes an unmanned aerial vehicle (UAV)-assisted low-altitude edge intelligent network, which aims to minimise the long-term cost by jointly optimising the offloading decisions of all tasks, resource allocation, and trajectory of all UAV-assisted MEC servers. To tackle this complex problem, we propose a soft actor-critic-based offloading and trajectory joint optimisation algorithm to seek the optimal resource allocation solutions in an MEC-enabled LAE network. The simulation results illustrate the convergence and effectiveness of the proposed algorithm, which are validated based on the comparison with benchmarks.