<p>For the real-time trajectory planning problem of speed-holding Automated Guided Vehicle (AGV) in the intelligent factory environment, an optimized trajectory planning algorithm based on the Frenet coordinate system is proposed. In this study, the trajectory optimization of AGV is taken as the research object, and the optimization objectives of selecting the optimal trajectory and improving the trajectory smoothness are taken as the optimization objectives. First, establishing the Frenet coordinate system spatial avoidance motion solves the complex problem that the relative position of AGV and lane change trajectory is challenging to represent in trajectory planning. Secondly, an artificial potential field method based on the Frenet coordinate system is proposed for trajectory planning, and the search cost and trajectory smoothness are optimized based on potential field and weighting. Finally, the loss function is constructed using the safety and smoothness evaluation metrics centered on the acceleration rate of change to evaluate the trajectory cost and combined with the curvature and acceleration checks to select the optimal solution that minimizes the loss. Algorithm validation experiments for AGV planning are conducted. The result is a search cost optimization of 24.34% and a maximum curvature optimization of 13.15% compared to the traditional artificial potential field. The result is a search cost optimization of 18.58% and a maximum curvature optimization of 5.71% compared to the traditional A* algorithm. The experimental results show that the algorithm can meet the planning requirements of the intelligent factory scenario with a smooth vehicle trajectory, more negligible cost, and higher safety.</p>