To address the challenges of inspecting multiple target road segments in large-scale transportation networks, where traditional manual methods are inefficient and single-drone operations are limited by endurance, this paper proposes a coordinated path planning and optimization method for multiple vehicles and multiple drones. A mixed-integer linear programming model is established with the objective of minimizing the total inspection time, considering constraints such as vehicle scheduling, task allocation, drone path planning, and time synchronization. An improved Teaching-Learning-Based Optimization (TLBO) algorithm is designed to solve the model effectively, incorporating a multi-dimensional solution structure and ten variable neighborhood search operators to enhance search capability and convergence precision in complex solution spaces. Experimental results demonstrate that the proposed algorithm outperforms traditional methods such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Iterated Local Search (ILS) across various large-scale test cases and real-world road networks, showing significant advantages in solution quality and stability.