Enhancing Battery Efficiency Through Semi-Markov Decision Processes in Task Allocation for UAVs

This research introduces a Semi-Markov Decision Process (SMDP) approach to task allocation for Unmanned Aerial Vehicles (UAVs) by considering the stochastic behavior of battery levels. The SMDP allows the UAV to dynamically allocate tasks based on its current battery state and the time spent in each state. We compare the SMDP method with a manually assigned task sequence and demonstrate a significant reduction in completion time for a set of predefined tasks. The SMDP optimally assigns tasks, considering the stochastic nature of battery levels, resulting in improved efficiency and eliminating uncertainties associated with human allocation. The findings underscore the benefits of incorporating SMDP in UAV task management, especially in scenarios dependent on real-time battery levels.