Deepwater robot multi-model EKF combined navigation device and method

Abstract

The invention relates to the technical field of deepwater robots, and especially relates to a deepwater robot multi-model EKF combined navigation method. A device related with the method comprises a range finder, a sound beacon, an inertial navigation sensor, and a navigation computer. The method is characterized in that the method comprises the steps: enabling the navigation computer to carry out the data collection from the range finder, the navigation sensor, a navigation speed sensor and a depth sensor after a deepwater robot enters the water; determining a navigation stage of the deepwater robiot according to the quantity and quality of distance measurement data; building a plurality of deepwater robot kinematic models when the deepwater robot is at the combined navigation stage, determining the corresponding weight of each model when the distance measurement data is updated, and solving the weighted output of all navigation models, wherein the weighted output is the output value of a combined navigation algorithm. The method is stable and reliable, is high in navigation precision in a deepwater environment, and is wide in application range. Moreover, the used device and method are simple, and are good in inheritance.

Description

technical field [0001] The invention relates to the technical field of deep-water robots, in particular to a multi-model EKF combined navigation device and method for deep-water robots. Background technique [0002] Deep-water robots can sail at a depth of several kilometers. During underwater navigation, the navigation error will gradually increase with the increase of distance and time. After the deep-water robot sails for a certain distance, its position needs to be corrected to improve the navigation accuracy. [0003] The current method of correcting the position of the deep-water robot is mainly to directly correct the position of the deep-water robot through the positioning information output by the acoustic positioning system. But this method has certain flaws: [0004] 1) The positioning error caused by the acoustic propagation delay cannot be eliminated; [0005] 2) There is a deviation in the calibration of the underwater robot, which will cause the navigation...

AI Technical Summary

Problems solved by technology

[0004] 1) The positioning error caused by the acoustic propagation delay cannot be eliminated;

[0005] 2) There is a deviation in the calibration of the underwater robot, which will cause the navigation deviation to accumulate over time;

[0006] 3) The navigation trajectory of the deep-water robot is not smooth;

[0007] 4) Unable to effectively overcome the interference of outliers

Images