To address the challenge of balancing efficiency and accuracy in measuring complex aero-engine blade surfaces, this paper proposes a robotic-arm trajectory planning method based on an improved multi-objective grey wolf optimization algorithm (ILMOGWO). The kinematic and dynamic models of a robotic arm-structured light hand-eye system are established, where end-effector vibration constraints are transformed into imaging clarity constraints according to structured-light principles. A multi-objective trajectory model minimizing total motion time and maximum vibration velocity is constructed using quintic polynomial interpolation. To enhance optimization performance, ILMOGWO integrates Latin hypercube sampling initialization, a nonlinear convergence factor, adaptive grid-based archive management, and Levy flight perturbation. Simulation and experimental results on a six degrees of freedom (6-DOF) industrial manipulator verify that the proposed method achieves superior Pareto front convergence and effectively suppresses end vibration within 0.115 mm/s. The reconstructed blade point cloud exhibits a maximum deviation of 0.050 9 mm and an average deviation of 0.016 0 mm. The proposed method can significantly improves the imaging quality of structured light and the accuracy of 3D reconstruction while ensuring measurement efficiency, providing a feasible trajectory optimization approach for high-precision measurement of complex surfaces.