Research on AI Electronic Fuel Injection of Agricultural Hybrid UAV

According to the problems existing in the EFI of the oil-electric hybrid UAV, this paper applies the Artificial Intelligence (AI) technology to the EFI, which reduces a lot of work of testing the EFI, improves the EFI efficiency and performance, and makes the oil-electric hybrid UAV have higher performance. The experimental data show that the error between the inference data and the actual test data of EFI is less than 8% after the AI neural network is trained, which shows that the trained neural network can meet the requirements of actual use, and can be used to replace MAP to obtain the optimal engine injection pulse width.