This paper proposes an Advanced Crayfish Optimization Algorithm (COAA) for UAV 3D path planning, aiming to address the limitations of traditional methods, including insufficient initial population diversity, imbalanced exploration and exploitation, and high susceptibility to local optima. The study begins by formulating the path planning problem as a multi-constrained optimization model, and it designs a multi-objective function to optimize path length, altitude variation, and smoothness constraints, while cubic spline interpolation ensures path feasibility. To overcome challenges of traditional algorithms in path planning, COAA introduces three key enhancements: (1) Tent chaotic mapping initialization to improve population diversity and solution space coverage; (2) an adaptive inertia weight mechanism to dynamically balance global exploration and local exploitation; and (3) a dynamic probability mutation strategy to strengthen local optimum escape capability. Comprehensive simulations were carried out in three distinct environments to systematically compare COAA with six typical algorithms. Results demonstrate that COAA achieves 4.3%–12% lower fitness values, highlighting its superiority in balancing exploration-exploitation efficiency and adapting to complex dynamic constraints, thereby showcasing the substantial potential for practical UAV mission planning applications.