Laser powered drones are simple to operate and with high safety,which use photovoltaic panels to receive energy for operation.Under complex shading conditions,the P V curve of the photovoltaic panel on the UAV has multiple extreme points.The traditional maximum power point tracking algorithm is easy to fall into the local optimum and the tracking speed is slow.Aiming at this problem,an improved bat algorithm is proposed.First,the population is initialized through reverse learning to enhance the uniformity of the population distribution and avoid falling into the local optimum.At the same time,the shrinkage factor is introduced to accelerate the algorithm′s convergence speed.Finally,a simulation model of the photovoltaic array circuit is built in Matlab/Simulink to prove that the improved bat algorithm has better speed and stability.