<p>To address the motion smoothness and rapid response requirements of automated hydrogen refueling for fuel cell electric vehicles (FCEVs), this study develops a robotic arm-based refueling scheme and focuses on a time-optimal trajectory planning method that minimizes the joint motion time under kinematic smoothness constraints. A six-degree-of-freedom (6-DOF) BRTIRUS1510A robotic arm is taken as the research object, and its kinematic model is established via the Denavit–Hartenberg (D–H) parameter method. A 6–5–6 polynomial interpolation function is proposed and solved to serve as the foundation for trajectory planning. To overcome the shortcomings of the conventional Seagull Optimization Algorithm (SOA), such as insufficient population diversity and premature convergence, a multi-strategy improved Seagull Optimization Algorithm (ISOA) is developed by incorporating Tent chaotic map initialization, a nonlinear control parameter strategy, a multi-directional adaptive position update strategy, and a Lévy flight strategy integrated with greedy selection. An effectiveness analysis of the improved strategies and a qualitative analysis are conducted. Furthermore, comparative experiments with other algorithms are carried out on 35 benchmark test functions, and the Wilcoxon rank-sum test and Friedman test are used for statistical evaluation. The results confirm that ISOA possesses superior optimization performance. When applied to the time-optimal trajectory planning of the robotic arm, the optimized joint motion time is reduced by 50.94%. The resulting kinematic curves are continuous and smooth, and the kinematic constraints are satisfied, verifying the effectiveness of the proposed method. The efficient and parallelizable algorithmic architecture of ISOA is suitable for deployment on high-performance computing platforms, offering technical support for large-scale real-time optimization scenarios.</p>