MIMU overall dynamic intelligent calibration compensation method

• Main Authors: ZHANG MENGJUN, CHENG YU, CHEN SHUAI, WANG BOHAO, MA YONGBEN, MA YIMING, BAI ZHENGHAO, ZHAO HAIFEI
• Format: Patent
• Language: Chinese; English
• Published: 09.07.2021
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; GYROSCOPIC INSTRUMENTS; MEASURING; MEASURING DISTANCES, LEVELS OR BEARINGS; NAVIGATION; PHOTOGRAMMETRY OR VIDEOGRAMMETRY; PHYSICS; SURVEYING; TESTING

The invention discloses an MIMU overall dynamic intelligent calibration compensation method. The method comprises the following steps of: fixing an MEMS-IMU on test equipment through a tool, and collecting and storing outputs of the IMU and the test equipment; establishing an error compensation model based on a DeepAR recurrent neural network, and determining a neural network input and output structure, the number of hidden layers and the number of neurons; taking the collected actual measurement data of the inertial measurement unit as the input of a neural network, respectively taking the difference between the data of the test equipment and the actual output of the inertial measurement unit as the output of a neural network model, establishing a training set and a test set, and training the neural network model by using the training set; and combining the trained neural network model with test set input to obtain output data of the model, comparing the output data with the output data of the test set, eval