The UAV/HAP-assisted Artificial Intelligence of Things (AIoT) integrates Artificial Intelligence (AI) technology into the Internet of Things (IoT) system, which provides ubiquitous global communications and intelligent computation services for mobile terminal devices (MTDs) by combining terrestrial and non-terrestrial communication networks. Efficient and energy-saving communication and computing resource support are crucial to improving the performance of AIoT systems. Therefore, an energy-efficient joint computation offloading and resource allocation strategy is proposed for AIoT supported by UAVs/HAPs. The strategy combines the advantages of Lyapunov optimization theory and the deep reinforcement learning (DRL)-based model to obtain optimal energy-efficient computation offloading and resource allocation decisions while guaranteeing long-term queue stability under dynamic network conditions. Firstly, the joint optimization problem is modeled as a multi-stage stochastic mixed integer nonlinear programming (MINLP) problem and reformulated into deterministic sub-problems under long-term performance constraints by the Lyapunov optimization theory. Then, a hybrid DRL-based online optimization algorithm is proposed to solve these sub-problems. This algorithm minimizes the overall system energy consumption by jointly optimizing computation offloading and resource allocation decisions. The simulation results show that the proposed strategy improves energy efficiency, decreases algorithm computation complexity, and guarantees the stability of task queues.

Integrating terrestrial and non-terrestrial networks can provide a wide area of Internet of Things (IoT) services with global connections and ubiquitous communications. However, integrated terrestrial and non-terrestrial networks (ITNTNs) also face huge challenges caused by complex environments, diverse services, and heterogeneous nodes. By leveraging the powerful driving force of edge intelligence (EI) technology for network development, a framework named “ITNTNs with EI” is proposed in this paper as the solution to the challenges above. We design system architecture, dynamic edge resource deployment, and intelligent edge model training for the proposed framework and discuss its application scenarios, challenges, and some open research issues. Then, simulation experiments about a joint computation offloading and resource allocation optimization algorithm based on the DRL-based model are conducted for a specific framework instance. The results demonstrate that optimization management solutions driven by EI can provide faster response and greater quality of service for IoT applications in ITNTNs.

UAV-based air-ground integrated networks offer a significant benefit in terms of providing ubiquitous communications and computing services for Internet of Things (IoT) devices. With the empowerment of edge intelligence (EI) technology, they can efficiently deploy various intelligent IoT applications. However, the trajectory of UAVs can significantly affect the quality of service (QoS) and resource optimization decisions. Joint computation offloading and UAV trajectory optimization bring many challenges, including coupled decision variables, information uncertainty, and long-term queue delay constraints. Therefore, this paper introduces an air-ground integrated architecture with EI and proposes a TD3-based joint computation offloading and UAV trajectory optimization (TCOTO) algorithm. Specifically, we use the principle of the TD3 algorithm to transform the original problem into a cumulative reward maximization problem in deep reinforcement learning (DRL) to obtain the UAV trajectory and offloading strategy. Additionally, the Lyapunov framework is used to convert the original long-term optimization problem into a deterministic short-term time-slot problem to ensure the long-term stability of the UAV queue. Based on the simulation results, it can be concluded that our novel TD3-based algorithm effectively solves the joint computation offloading and UAV trajectory optimization problems. The proposed algorithm improves the performance of the system energy efficiency by 3.77%, 22.90%, and 67.62%, respectively, compared to the other three benchmark schemes.