Improved ant colony algorithm-based research on UAV trajectory planning

Based on the two main shortcomings of the classical ant colony algorithm in UAV trajectory planning: first, the lack of directional guidance for the initial search, the problem of high blindness, and second, the ease of falling into the local optimum, a new pheromone updating function and mechanism are proposed, whose basic idea is that: ants with the shortest search paths of a generation are recorded as the searched paths, and compared with the optimal paths of the current historical records. The pheromone concentration of the intersection part will be enhanced appropriately. The theoretical analysis and simulation experiments are carried out on the optimization strategy, and the feasibility, superiority and stability of the improved algorithm in the trajectory planning problem are verified through several repeated experiments. The results of simulation experiments show that the improved ACO algorithm proposed in this paper reduces the average path length by 7.2% and the average number of iterations by 40.3% compared with the classical ACO algorithm for trajectory planning, which verifies that the present algorithm is able to effectively improve the stability of UAV trajectory planning.