Objective To further enhance the intelligence level of food production, and address the issues of insufficient flexibility and operation accuracy of existing Delta robots for food sorting in applications, thus improving the overall efficiency of food production lines and product quality stability.Methods Based on the intelligent food production system, a trajectory optimization and tracking method for intelligent food production robots is proposed, which integrates multi-objective trajectory optimization and adaptive fuzzy sliding mode control (AFSMC). Firstly, a multi-objective (energy consumption and operation time) optimization model is built. Next, the model is solved through an improved particle swarm algorithm to generate the optimal motion trajectory that balances efficiency and energy consumption. Then, real-time tracking control of the optimized trajectory is performed through fuzzy adaptive control to ensure the motion accuracy of the robot in complex working conditions. Finally, the performance of the proposed method is validated by a food sorting experimental platform.Results Compared with traditional trajectory control methods, the proposed method reduces the energy consumption by >3% and shortens the single operation time by >3%, with a trajectory tracking error <0.5 mm. Additionally, the proposed model can maintain stable operating accuracy and flexibility in complex scenarios, such as different food shapes and conveyor belt speed.Conclusion Through the collaborative application of multi-objective trajectory optimization and fuzzy adaptive control, the proposed method effectively addresses the issues of low flexibility and accuracy of existing food production robots, significantly reducing energy consumption and operation time.