Motion-Encoded Electric Charged Particles Optimization for Moving Target Search Using Unmanned Aerial Vehicles

Area of Science:

• Robotics and Control Systems
• Optimization Algorithms
• Artificial Intelligence

Background:

• Locating moving targets with unmanned aerial vehicles (UAVs) presents significant challenges.
• Existing optimization algorithms require enhancement for dynamic tracking applications.
• Bayesian theory provides a framework for probabilistic target localization.

Purpose of the Study:

• To introduce a novel optimization algorithm, ECPO-ME, for detecting moving targets using UAVs.
• To improve the probability of target detection by formulating it as an optimization problem.
• To evaluate the performance of ECPO-ME against established metaheuristic algorithms.

Main Methods:

• Developed the motion-encoded electric charged particles optimization (ECPO-ME) algorithm by integrating a motion encoding (ME) mechanism with the ECPO algorithm.
• Formulated the moving target search as an optimization problem maximizing detection probability using Bayesian theory.
• Encoded UAV search trajectories as evolving motion paths within the ECPO-ME iterations.
• Tested ECPO-ME across six diverse scenarios and performed statistical comparisons with other metaheuristics.

Main Results:

• The ECPO-ME algorithm demonstrated superior performance in identifying moving targets across all tested scenarios.
• ECPO-ME outperformed the base ECPO algorithm by significant margins, ranging from 0.79% to 14.72% across different scenarios.
• Statistical analysis confirmed the effectiveness and robustness of ECPO-ME compared to other benchmarked metaheuristics.

Conclusions:

• The proposed ECPO-ME algorithm shows substantial potential for real-world applications such as tracking misplaced animals.
• ECPO-ME offers an effective solution for enhancing the accuracy and efficiency of moving target detection using UAVs.
• The integration of motion encoding significantly improves the performance of the electric charged particles optimization algorithm for dynamic tracking.