| CPC B25J 9/1664 (2013.01) [G06T 7/73 (2017.01); G06V 10/751 (2022.01); G06V 20/50 (2022.01); H02J 7/00032 (2020.01); H04B 17/318 (2015.01)] | 7 Claims |
|
1. A recharging control method of a desktop robot, the desktop robot comprises a robot body provided with a camera and a base integrating charging and wireless beacon transmitting modulating functions, wherein the recharging control method comprises:
step A, when the robot body of the desktop robot receives a charging instruction of a recharging mode, calling the camera to collect images of a desktop environment, controlling two continuous frames of the images to be subjected to feature matching, calculating position change of the robot body, and then constructing a corresponding grid map;
step B, controlling the robot body of the desktop robot to receive a wireless beacon modulated and transmitted by the base, and analyzing initial orientation of the base according to strength information of the wireless beacon;
step C, planning a recharging path in the grid map in combination with the position change of the robot body and the initial orientation obtained through analysis; and
step D, controlling the robot body of the desktop robot to move along the recharging path;
step E, when the strength of the wireless beacon is within a preset threshold range, controlling the collected two continuous frames of images to be subjected to feature matching, then calculating a vector value of current position of the robot body, and further searching the base according to the vector value;
wherein a signal source for transmitting the wireless beacon is arranged on the base, and the preset threshold range is 80%-90% of the maximum value of the signal strength of the wireless beacon transmitted by the signal source; and the two continuous frames of images are images of the same object on imaging planes of the camera at different visual angles;
wherein in the step A and the step E, the feature matching comprises: selecting center points of the images collected by the camera to serve as feature points of environment images carrying out feature descriptor matching with the environment image in a desktop map database built in the desktop robot, so as to find the images with the maximum matching, the desktop map database stores environment image feature descriptors corresponding to all positions on a recharging circumference and a circle center position, and the recharging circumference takes the center position of the base as the circle center and a predetermined length as a radius on the desktop where the robot body moves, and desktop coordinate values measured in advance in an actual environment; wherein the predetermined length has the value of no more than 20 cm,
wherein in step E, calculating a vector value of the current position of the robot body comprises: after the images with the maximum matching degree are obtained in the feature matching process of the collected continuous two frames of images, calculating a homography matrix of the collected continuous two frames of images according to the feature points of the images with the maximum matching degree and an internal parameters of the camera; according to constraint relation corresponding to the homography matrix, unitizing the coordinate values of the feature points of the collected continuous two frames of images into the same coordinate system; and calculating distance between the robot body of the desktop robot and the corresponding position on the recharging circumference when the current environment image is acquired according to a pinhole imaging principle and the coordinate values in combination with the homography matrix and the internal parameters of the camera, and further calculating the vector value of the current position of the robot body.
|