Recently, the use of unmanned aerial vehicles (UAVs) for gathering information from distributed sensors has emerged as a key focus in Internet of Things (IoT). While previous studies primarily concentrated on minimizing acquisition time, moderating power utilization, and for increasing volume of gathered information, they often overlooked the optimization of data freshness. The primary goal of this research is to enable UAVs to achieve long-term data collection tasks in dynamic environments while managing the age of information (AoI) and adhering to their power limitations. To tackle these complexities, this approach aims to maximize the Quality of Service (QoS) by utilizing the CAViaR Eel and Grouper Optimizer (CaViaREGO) for trajectory generation in UAV-IoT. Firstly, system model is simulated. After that, range constraints and collision avoidance among UAVs is performed. Then, trajectory is generated using CaViaREGO by considering fitness factors, such as throughput, delay, energy, and distance. CaViaREGO is the combination of Eel and Grouper Optimizer (EGO) with Conditional Autoregressive Value at Risk by Regression Quantiles (CAViaR). The evaluation of presented scheme resulted in optimal values for path length, energy consumption, speed, and fitness metrics of 9.825, 0.693 J, 26.502 m/s, and 0.824, respectively.