Improved ant colony algorithm with Q-learning adaptation mechanism for coverage UAV path planning in search problem

Unmanned aerial vehicles (UAVs) are widely used in search problem due to their portability and high performance. In UAV-assisted search problem, the path planning is considered as the coverage path planning problem, which is usually converted to a traveler's problem through the grid decomposition method. To solve this problem, this paper has designed an improved ant colony algorithm, which combines Q-learning based adaptive strategy, elite strategy and other methods to enhance the exploration and convergence ability. Simulation results show that the method can effectively improve the coverage efficiency of the multi-UAV multi-area coverage search problem and reduce the UAV flight energy consumption.