Underwater robot underwater positioning and path planning method based on sonar image processing

Abstract

The invention provides an underwater robot underwater positioning and path planning method based on sonar image processing. The underwater robot underwater positioning and path planning method comprises the following steps that obtaining sonar images, direction information, position information and geometrical information of an underwater structure; obtaining the operation condition of an underwater robot in the underwater structure; obtaining the current position of the underwater robot according to the sonar images, the direction information, the position information and the geometric information of the underwater structure and the operation condition of the underwater robot; constructing an occupied grid chart according to the direction information, the position information and the geometric information of the underwater structure; detecting an obstacle encountered by the underwater robot during operation in the underwater structure according to an original image and the occupied grid chart; and planning an operation path of the underwater robot according to the occupied grid chart and the obstacle. According to the method, the problems of measurement deviation influence and underwater noise interference on the underwater robot in the underwater positioning and path planning process can be reduced, so that the accuracy and adaptability of underwater positioning and path planning of the underwater robot are improved.

Description

technical field [0001] The invention relates to the technical field of automatic positioning and automatic navigation, in particular to an underwater positioning and path planning method for an underwater robot based on sonar image processing. Background technique [0002] Autonomous underwater vehicle navigation generally utilizes inertial measurement units, attitude and heading reference systems, and Doppler velocimeters for inertial navigation, but these sensors are susceptible to bias, resulting in low navigation accuracy. [0003] Autonomous navigation methods for underwater robots are currently based on the use of vision sensors or sonar sensors, where vision-based algorithms cannot handle localization tasks near installed subsea structures because the algorithms require a specific configuration of the target object, such as additional fiducial markers, Therefore, the vision-based algorithm is less adaptable; moreover, the sonar-based algorithm is susceptible to bias, ...

AI Technical Summary

Problems solved by technology

For this reason, the purpose of the present invention is to propose a method for underwater positioning and path planning of underwater robots based on sonar image processing, which can reduce the influence of measurement deviation and underwater The problem of noise interference, thereby improving the accuracy and adaptability of underwater positioning and path planning of underwater robots

Images