Abstract: Parameter tuning is a critical step in PID controller design. The traditional method relying on manual experience is time-consuming and labor-intensive, and it is prone to falling into local optima when applied to UAV attitude tracking control. Therefore, it is of great significance to introduce optimization algorithms for automatic parameter tuning. This paper proposes an improved sparrow search algorithm (ISSA) to reduce the possibility of the original sparrow search algorithm (SSA) falling into local optima, enhance the global ooptimization capability, and thus effectively solve the problems of low efficiency and manual dependence in the tuning process. In ISSA, firstly, a hybrid perturbation strategy is used to initialize the population, strengthening global exploration ability. Secondly, adaptive weights, Cauchy perturbation, and Lévy flight are introduced into the discoverer phase to improve search flexibility and adaptability. Finally, a dynamic scent strategy is adopted in the tracking phase to enhance local search accuracy and convergence speed. The feasibility and practical value of the proposed algorithm are validated through parameter tuning experiments for UAV attitude tracking control.