In response to the challenge of parallel processing of multiple tasks (workpieces) in carrier-based air-craft support operations, which are abstracted as a flexible flow workshop scheduling problem, and the limitations of existing research in the collaborative scheduling of heterogeneous carrier-based aircraft, a dynamic parallel scheduling method that integrates a central scheduling mechanism with a deep reinforcement learning decision model is proposed. Initially, the parallel time series of support operations is equivalently transformed into a serial logical sequence. This transformation ensures compatibility with the flexible flow workshop scheduling model while preserving the characteristic of parallel execution. Subsequently, a Markov model for job scheduling decisions is constructed based on the logical sequences, incorporating the operational differences between manned and unmanned aerial vehicles. Distinct decision models are designed and trained for each type of aircraft. Moreover, a central scheduling mechanism is developed to unify the management of these two decision models, coordinating global positioning, resources, and other situational information. This mechanism disseminates information to the respective decision models to facilitate effective collaboration. Finally, simulation comparison experiments indicate that the proposed algorithm significantly enhances decision real-time performance, even at the cost of marginal scheduling efficiency, compared to optimization methods represented by genetic algorithms. The algorithm effectively balances carrier-based aircraft deployment time and the output time of scheduling methods, making it particularly suitable for rapid deployment tasks in high-real-time and dynamic environments.

A joint optimization method for unmanned aerial vehicle (UAV) trajectory planning and resource allocation based on deep reinforcement learning was proposed to address the challenges of limited battery capacity, limited cache space, and dynamic changes in ground target priorities during data collection tasks in emergency scenarios. First, a mathematical model was developed by considering the communication, computation, flight, and data caching processes in UAV missions. Then, a Markov process model was established for UAV trajectory planning and resource allocation, with corresponding state and action descriptions. A weighted reward function was designed to balance UAV energy consumption and data collection volume. Finally, simulations were conducted to compare the proposed method with greedy algorithms and genetic algorithms. The results show that the proposed method can significantly improve the amount of data collected from ground users within a shorter task time, at a similar or lower energy cost for UAVs.