| 摘要 |
PURPOSE:To accurately make magnetic corrections by using a learning coefficient for a last azimuth circle center as to a calculated azimuth circle center, performing learning operation by a specific ratio, and obtaining a new azimuth circle center. CONSTITUTION:Output data which is detected by an earth magnetism azimuth sensor 1 and based upon the earth magnetism, output data on the azimuth change of a vehicle from the difference between the right and left wheel speeds of the vehicle which is detected by a wheel speed sensor 2, and output data on the moving distance of the vehicle which is detected by a travel distance sensor 3 are inputted to an arithmetic part 5 through a sensor interface 5. Then the arithmetic part 5 calculates the absolute azimuth and the azimuth variation quantity of the vehicle from the output data of the sensors 1 and 2 and evaluates the disorder of XY coordinates and the disorder of the absolute azimuth by using the radius value of the azimuth circle which is learnt so far as an evaluation reference value according to past several data including individual current values. Further, the radium values of the evaluation reference value which are learnt so far are further learnt and made to correspond to variation in the solarized quantity of the vehicle body. |
