Automobile front axle structure is one of the core components of an automobile, which plays a decisive role in the automobile design. In order to improve the convergence speed and accuracy of the automobile front axle lightweight, an improved manta ray algorithm based on manta ray self-defense strategy is proposed. Six classical benchmark functions are used to verify the performance of the improved manta ray algorithm. Results show that the improved manta ray algorithm has competitive convergence speed and convergence accuracy. On this basis, the improved manta ray algorithm is used to optimize the lightweight design of the automobile front axle. The optimization results show that the optimal solution can be obtained after 94 iterations, and the total mass of the automobile front axle after optimization is reduced from 51.95 kg to 43.24 kg, a decrease of 16.75%. By analyzing the results of classical benchmark functions and automobile front axle cases, it can be seen that the improved manta ray algorithm is an efficient optimization algorithm, which provides reference for future engineering optimization problems and algorithm improvements.