A UAV path planning algorithm for bridge construction safety inspection in complex terrain

In response to the challenge of rapid unmanned aerial vehicles (UAV) path planning for bridge construction in complex terrain, this paper presents an enhanced snake optimization (CSGLSO) UAV three-dimensional path planning algorithm. Initially, this study enhances the stochasticity strategy for generating initial populations within the Snake Optimization (SO) algorithm employing the Piecewise Chaotic Mapping technique, thereby obliterating transient periodic traits and fostering equilibrium in the solution space of the SO algorithm's progenies. Subsequently, integrating the Subtraction-Average-Based Optimizer algorithm mitigates the issue of convergence speed within the SO algorithm confronting high-dimensional complex functions. Ultimately, employing adaptive t-distribution and lens imaging reverse learning facilitates the evasion of local optima within the current position by the SO algorithm, thus augmenting its exploratory prowess. To ascertain the efficacy of the enhanced algorithm, 14 standard test function convergence comparison experiments were conducted, as well as three-dimensional path planning simulation experiments under multi-scenario conditions of bridge construction by UAV. Experimental findings reveal that relative to SO, Hybrid Snake Optimizer Algorithm, Improved Salp Swarm Algorithm, and Exploratory Cuckoo Search, CSGLSO manifests shorter and more streamlined trajectories, accelerated convergence rates, and elevated optimization precision. Thereby, UAVs are empowered to execute path-planning endeavors expeditiously and precisely within intricate environments.