Abstract: This paper proposes variable step size chaotic firefly algorithm (VS-CLSFA) and immune particle swarm optimization (IM-PSO), comparing particle swarm optimization (PSO), firefly algorithm (FA) and improved firefly algorithm (MFA) to the tracking results of the maximum power point of photovoltaic modules in solar UAV under partial shadow. At the same time, the tracking effects of the above algorithms under the influence of factors such as the flying height and speed of the solar UAV is studied. The results show that VS-CLSFA and IM-PSO have overcome the shortcomings of FA, MFA and PSO that are trapped in local optimal or premature convergence, and quickly and stably track the maximum power point of photovoltaic modules in solar UAV. For photovoltaic modules with more complex output characteristics, the above five algorithms need to increase the number of iterations and sacrifice tracking time to improve tracking accuracy and stability; Compared with VS-CLSFA, the tracking accuracy of IM-PSO increases by about 0.229% , and tracking time reduces by about 0.108 s.