A mobile ad hoc network (MANET) comprises multiple autonomous unmanned aerial vehicles (UAVs) connected in an ad hoc manner. MANETs are key components in achieving different services in smart cities. One of the main issues in the MANET is UAV placement, which refers to finding the optimal positions of UAVs. Recently, researchers proposed several machine-learning methods for UAV placement. The existing techniques obtained promising results; however, their performance is far from the best, and more effort is needed. This paper introduces an adaptive local search-based arithmetic optimization (LSAO) algorithm for UAV placement. The incentive mechanism of LSAO is enhancing the search dynamics by embedding an adaptive switching probability, a chaotic local search, and an opposition-based learning strategy into the standard AO algorithm. The proposed method is benchmarked on well-known placement test cases, and the results are verified by a comparative study with state-of-the-art algorithms. The results confirm that LSAO generated competitive outcomes compared to its peers in most simulation benchmarks. The LSAO obtained the first rank in terms of coverage, connectivity, and total fitness values among comparison algorithms.
