Research and Design of Five-Axis Balanced Camera Stabilizer Based on BP Neural Network PID Algorithm

This paper adaptively adjusts the three parameters of proportional, integral and differential for five-axis balanced camera stabilizer. Real-time adjustment of the input of the controlled object, so that the system responds quickly and stabilizes, thereby reducing the blur caused by external factors such as jitter caused by the camera's captured image. Enables a handheld camera stabilizer to take a clear picture while the photographer's arm shakes. In order to improve the stability of the triaxial hand-held camera, the wobble of the up and down motion cannot be eliminated. Two mechanical anti-jitter shaft arms are loaded under the triaxial stabilizer to make it a five-axis stabilizer, which can keep the camera picture stable in the pitching roll course and five directions of movement above and below. Through the corresponding simulation experiment, it is verified that the control of PID controller through BP neural network can make the system have high accuracy and strong stability, maintain the balance of the picture, and obtain accurate control effect for the stable position of the camera.