Positioning method, device and equipment

Abstract

The embodiment of the invention provides a positioning method, device and equipment, and the method, device and equipment are applied to electronic equipment, and the method comprises the steps: respectively obtaining the poses, detected by an inertial sensor when an image collector collects a current frame and a reference frame, of the electronic equipment; determining a relative pose that minimizes the sum of the luminosity error between the reference frame and the current frame and the relative motion error detected by the inertial sensor; obtaining the pose of the electronic equipment at the current frame by utilizing the pose and the relative pose corresponding to the reference frame; obtaining the optical flow of the current frame; and under the condition that the optical flow is greater than a preset threshold, taking the current frame as a new key frame, optimizing the pose of each key frame in a manner of minimizing the sum of a luminosity error and a relative motion error between two adjacent key frames by utilizing the previously stored key frames, and taking the new key frame subjected to pose optimization as a new reference frame. By applying the scheme provided by theembodiment of the invention, the robustness of a positioning result in an extreme environment can be improved.

Description

technical field [0001] The present invention relates to the technical field of positioning and navigation, in particular to a positioning method, device and equipment. Background technique [0002] Autonomous positioning is the core component of the robot's autonomous navigation system. On the basis of autonomous positioning, the robot can realize functions such as obstacle avoidance and autonomous navigation. [0003] In the prior art, the robot can use the visual odometry to carry out autonomous positioning, which specifically includes: using the camera installed on the robot to capture images in real time and estimating its own position through the visual odometry (including the direct method odometry and the feature point method odometry). Pose, that is, to obtain its own position and posture, so as to complete autonomous positioning. [0004] Although in the prior art, the robot can use visual odometry for autonomous positioning, but because the camera is easily affect...

AI Technical Summary

Problems solved by technology

[0004] Although in the prior art, the robot can use visual odometry for autonomous positioning, but because the camera is easily affected by factors such as illumination changes, moving objects, camera occlusion, and less-textured scenes when taking images, it is difficult to estimate Deviation or even complete loss of own pose

That is to say, when the robot uses the above method for autonomous positioning, the positioning result is affected by various extreme factors in the above environment, and the robustness is poor.

Images