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

Area of Science:

• Robotics and Automation
• Artificial Intelligence
• Computational Optimization

Background:

• Unmanned aerial vehicles (UAVs) face challenges in rapid path planning for complex tasks like bridge construction.
• Existing Snake Optimization (SO) algorithms struggle with convergence speed and local optima in high-dimensional problems.

Purpose of the Study:

• To develop an enhanced 3D path planning algorithm for UAVs in complex terrain.
• To improve the efficiency, speed, and precision of UAV path planning using an optimized SO algorithm.

Main Methods:

• Enhanced Snake Optimization (CSGLSO) algorithm incorporating Piecewise Chaotic Mapping for initial population stochasticity.
• Integration of Subtraction-Average-Based Optimizer to address SO's convergence speed issues.
• Adaptive t-distribution and lens imaging reverse learning to enhance SO's exploration and avoid local optima.

Main Results:

• CSGLSO demonstrated superior performance in convergence speed and optimization precision compared to SO, Hybrid Snake Optimizer, Improved Salp Swarm Algorithm, and Exploratory Cuckoo Search.
• Simulations showed CSGLSO generates shorter, more streamlined trajectories for UAVs.
• The algorithm effectively navigated complex 3D environments for bridge construction scenarios.

Conclusions:

• The CSGLSO algorithm significantly enhances UAV path planning capabilities for complex tasks.
• CSGLSO offers a robust solution for rapid and precise navigation in challenging environments.
• This advancement empowers UAVs for more efficient bridge construction operations.