Abstract: An auxiliary force control system and a method for a power-assisted bicycle are disclosed. The system has a sensing device, a mobile computing device, a first controller, and a second controller. The sensing device receives a riding torque and a riding speed. The mobile computing device generates a tuning factor via a first and a second ANN model. Personal data and historical riding data are input data of the first ANN model. Predicted grade outputted by the first ANN model, the personal data, and environment data are input data of the second ANN model. The first controller generates a final factor according to the tuning factor, a mode factor, and a gap-range factor. The second controller outputs a motor driver current according to a parameter of target output of the motor, which is generated based on the final factor, to the motor to drive the motor.