<![CDATA[Stabilizing a camera mounted on an unmanned aerial vehicle (UAV) is critical for obtaining high-quality aerial imagery, particularly under dynamic disturbances and rapid maneuvers. Conventional proportional-integral-derivative (PID) controllers are widely employed in gimbal systems due to their simplicity, yet their performance strongly depends on precise parameter tuning. Classical methods, such as Ziegler–Nichols, often result in excessive overshoot and slow settling time, which are unsuitable for high-precision applications. This study introduces a metaheuristic-driven approach based on Ant Colony Optimization (ACO) integrated with a multi-objective cost function to tune PID parameters for a three-axis UAV camera gimbal system. The cost function simultaneously considers rise time, settling time, overshoot, and steady-state error, providing a balanced performance across all dynamic response metrics. A dynamic model of the gimbal actuators was developed, and simulations were performed in MATLAB using identical initial conditions derived from Ziegler–Nichols tuning. Comparative experiments demonstrate that the proposed method significantly outperforms classical tuning, reducing overshoot from 47.6% to 0% and improving settling time from 4.54 s to 1.28 s on the roll axis, with similar improvements on pitch and yaw. These results highlight the effectiveness of multi-objective ACO for high-precision stabilization, offering a promising direction for real-time UAV control systems.]]>
