Multi-mission UAV Trajectory Planning in Smart Agriculture with Polarization Learning Model-Driven by Harris Hawks Optimizer

With the aim of enhancing the capability of UAVs to perform multi-mission trajectory planning in smart agriculture, a polarization model-based Harris Hawk Optimizer (PL-HHO) is proposed. Within the algorithm component, the algorithmic diversity in the trajectory search process is greatly enhanced by the introduced polarization model, which prompts the algorithm to maximally traverse more trajectory nodes in the search space. Meanwhile, the nonlinear mechanism and random spiral perturbation model are introduced to help the algorithm to improve the search ability in the smart agriculture environment and avoid falling into the local optimal region. The results in the CEC2015 test function and two real scenarios of flight mission experiments show that PL-HHO not only has excellent theoretical optimization searching ability, but also can find low-cost and high-quality trajectory planning paths in different scales of backgrounds, which provides a reliable guarantee for the development of smart agriculture.