Provided is an unmanned vehicle real-time posture measurement method, comprising the following steps: (1) selecting a Kalman filter, and collecting raw data from a micro-inertia measurement system; (2) defining a measurement vector, and obtaining a calculated measurement vector according to a formula; (3) calculating a model error vector; (4) estimating a posture quaternion by means of a least square error standard function; (5) using a regression matrix to rotate the quaternion, calculating an optimal quaternion related to a measured acceleration and geomagnetic field in a body coordinate system, and taking the optimal quaternion as a measurement value of the Kalman filter; (6) setting a constraint condition, obtaining a reduced order matrix of a linear matrix, and thereby obtaining a posture parameter of an unmanned vehicle.
展开▼
