Autonomous Unmanned Aerial Vehicles (AUAVs) use software, rather than human pilots, to decide where and when to fly, when to capture data, and when to land. AUAVs generate detailed and dynamic maps that inform decisions across a wide range of applications from digital agriculture to wildlife conservation to forestry to smart cities. For example, in digital agriculture, farmers use crop health maps to tailor the application of pesticides to the specific needs of each management zone, increasing crop yield and optimizing resources. AUAVs use edge computing to process captured data and determine their flight path. However, edge processing demands differ between competing designs for AUAVs. For example, AUAVs that fly to preset waypoints in an automated fashion consume significant battery resources but fewer computational resources. In contrast, reinforcement learning (RL) designs wherein AUAVs select waypoints to maximize their reward function can save battery but require more computational resources. This poster will discuss our early efforts and research strategies in profiling the trade-offs in battery and computational resources from automated and RL approaches for zoom maneuvers. A pivotal element for low-cost mapping, zoom maneuvers reduce the AUAV's altitude to increase data resolution, capturing details previously obscured at higher altitudes. Zoom maneuvers can qualitatively improve the efficacy of AUAV missions, impacting resource requirements. In practice, a better understanding of zoom approaches will provide additional avenues for optimization of map generation, improving the ability of autonomous AUAVs to scale to large scale missions.