To address the issues of insufficient optimization precision, low optimization efficiency, and susceptibility to local optima when the Sparrow Search Algorithm (SSA) is applied to unmanned vehicle path planning, a multi-strategy improved SSA is proposed. The initial population is initialized using a quadratic composite chaotic mapping to enhance population diversity. An adaptive inertia weight mechanism based on the Logistic function is incorporated into the position update of the discoverers to balance the global and local search capabilities of the algorithm. The position update of the followers is improved by introducing a nonlinear perturbation factor in the Sine Cosine Algorithm, which enhances the optimization speed and precision and improves the ability to escape local optima. Experimental results demonstrate that the improved SSA achieves significant optimization in both precision and efficiency, validating its effectiveness in unmanned vehicle path planning.