CN-121973210-A - Robot control method and device, controller and robot

CN121973210ACN 121973210 ACN121973210 ACN 121973210ACN-121973210-A

Abstract

The application relates to a robot control method, a robot control device, a controller and a robot. The method is applied to a robot, the robot is provided with a depth camera and a wide-angle camera, an acquisition area of the two cameras has an intersection area, the acquisition area of the wide-angle camera covers at least part of an acquisition blind area of the depth camera, the method comprises the steps of determining a first position of an obstacle based on a depth image when the obstacle is determined to be in the intersection area based on the depth image in the moving process of the robot, determining a second position of the obstacle in the wide-angle image according to the first position, and performing movement control on the robot based on the wide-angle image and the second position when the obstacle is determined to enter the depth acquisition blind area, wherein the depth acquisition blind area is the acquisition blind area of the depth camera covered by the acquisition area of the wide-angle camera, and the accuracy of robot control can be improved by adopting the method.

Inventors

LIU XIAOLONG
ZHANG CAO
LI YANG

Assignees

重庆凤凰技术有限公司

Dates

Publication Date: 20260505
Application Date: 20260203

Claims (10)

1. The robot control method is characterized by being applied to a robot, wherein the robot is provided with a depth camera and a wide-angle camera, an intersecting area exists between an acquisition area of the depth camera and an acquisition area of the wide-angle camera, and the acquisition area of the wide-angle camera covers at least part of an acquisition blind area of the depth camera, and the method comprises the following steps: determining a first position of an obstacle based on a depth image acquired by the depth camera when the obstacle is determined to be in the intersection area based on the depth image acquired by the depth camera in the moving process of the robot; and under the condition that the obstacle is determined to enter a depth acquisition blind area, carrying out movement control on the robot based on the wide-angle image acquired by the wide-angle camera and the second position, wherein the depth acquisition blind area is the acquisition blind area of the depth camera covered by the acquisition area of the wide-angle camera.
2. The method of claim 1, wherein the performing movement control for the robot based on the wide-angle image acquired by the wide-angle camera and the second position comprises: Determining a wide-angle image area corresponding to the obstacle in the wide-angle image based on the second position; performing obstacle tracking in the wide-angle image acquired by the wide-angle camera based on the image characteristics of the wide-angle image area corresponding to the obstacle to obtain an obstacle tracking position; And performing movement control on the robot based on the obstacle tracking position.
3. The method of claim 2, wherein the performing movement control for the robot based on the obstacle tracking location comprises: and controlling the robot to stop moving under the condition that the distance between the robot and the obstacle is determined to meet a first preset distance condition based on the obstacle tracking position.
4. A method according to claim 3, wherein after said controlling said robot to stop moving, said method further comprises: And in a preset time period, performing obstacle tracking based on wide-angle images continuously acquired by the wide-angle camera, and controlling the robot to restart moving under the condition that the distance between the robot and the obstacle is determined to meet a second preset distance condition.
5. The method according to claim 1, wherein the method further comprises: And under the condition that the obstacle is determined to enter a depth acquisition blind area and the obstacle is determined to have a movement trend away from the acquisition area of the wide-angle camera, controlling the wide-angle camera to rotate based on the movement trend so as to enable the obstacle to be positioned in the acquisition area of the wide-angle camera.
6. The method of claim 5, wherein the controlling the wide-angle camera to rotate based on the motion trend comprises: determining a rotation direction of the wide-angle camera based on a movement direction included in the movement trend; determining a rotational speed of the wide-angle camera based on a movement speed included in the movement trend; And controlling the wide-angle camera to rotate based on the rotation direction, the rotation speed and a preset rotation limiting range.
7. The method of claim 1, wherein determining a second position of the obstacle in the wide-angle image acquired by the wide-angle camera based on the first position comprises: Determining a position conversion relation based on camera external parameters and camera internal parameters corresponding to the wide-angle camera and the depth camera respectively; and carrying out coordinate mapping on the first position based on the position conversion relation to obtain a second position of the obstacle in the wide-angle image acquired by the wide-angle camera.
8. The robot control device is characterized by being applied to a robot, wherein the robot is provided with a depth camera and a wide-angle camera, an intersecting area exists between an acquisition area of the depth camera and an acquisition area of the wide-angle camera, and the acquisition area of the wide-angle camera covers at least part of an acquisition blind area of the depth camera, and the device comprises: The detection module is used for determining a first position of the obstacle based on the depth image acquired by the depth camera when the obstacle is in the intersection area based on the depth image acquired by the depth camera in the moving process of the robot; the switching module is used for determining a second position of the obstacle in the wide-angle image acquired by the wide-angle camera according to the first position; and the control module is used for carrying out movement control on the robot based on the wide-angle image acquired by the wide-angle camera and the second position under the condition that the obstacle is determined to enter a depth acquisition blind area, wherein the depth acquisition blind area is the acquisition blind area of the depth camera covered by the acquisition area of the wide-angle camera.
9. A controller comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any one of claims 1 to 7 when the computer program is executed.
10. A robot comprising the controller of claim 9, a robot body, and a depth camera and a wide-angle camera disposed on the robot body; an intersection area exists between the collection area of the depth camera and the collection area of the wide-angle camera, and the collection area of the wide-angle camera covers at least part of a collection blind area of the depth camera.

Description

Robot control method and device, controller and robot Technical Field The present application relates to the field of robot control technologies, and in particular, to a method and apparatus for controlling a robot, a controller, and a robot. Background With the development of robot technology, the autonomous recognition of the obstacle of the robot is a key for realizing the autonomous control of the robot, and the robot can automatically control the running state, such as stopping movement or restarting movement, according to the recognized obstacle. In the related art, obstacle recognition is usually realized by using a depth camera mounted on a robot, but when an obstacle enters the camera blind area, the obstacle position is usually predicted by using the motion parameters of the obstacle, and the robot is controlled according to the predicted obstacle position, so that the problem of lower control accuracy of the robot exists. Disclosure of Invention In view of the above, it is necessary to provide a robot control method, a robot control device, a controller, and a robot capable of improving the control accuracy of the robot. In a first aspect, the present application provides a robot control method applied to a robot, the robot is provided with a depth camera and a wide-angle camera, an acquisition area of the depth camera and an acquisition area of the wide-angle camera have an intersection area, and the acquisition area of the wide-angle camera covers at least part of an acquisition blind area of the depth camera, the method includes: determining a first position of an obstacle based on a depth image acquired by the depth camera when the obstacle is determined to be in the intersection area based on the depth image acquired by the depth camera in the moving process of the robot; and under the condition that the obstacle is determined to enter a depth acquisition blind area, carrying out movement control on the robot based on the wide-angle image acquired by the wide-angle camera and the second position, wherein the depth acquisition blind area is the acquisition blind area of the depth camera covered by the acquisition area of the wide-angle camera. In one embodiment, the moving control of the robot based on the wide-angle image acquired by the wide-angle camera and the second position includes: Determining a wide-angle image area corresponding to the obstacle in the wide-angle image based on the second position; performing obstacle tracking in the wide-angle image acquired by the wide-angle camera based on the image characteristics of the wide-angle image area corresponding to the obstacle to obtain an obstacle tracking position; And performing movement control on the robot based on the obstacle tracking position. In the embodiment, the wide-angle image area corresponding to the obstacle is determined through the second position, the image features of the area are extracted to track, and finally the movement control is performed on the robot according to the tracking position, so that the obstacle area in the wide-angle image can be accurately framed based on the second position, blind analysis on the whole wide-angle image is not needed, the occurrence of false recognition is improved, the accuracy of the tracking position is ensured, the tracking is performed by depending on the image features of the obstacle area, the appearance and the movement characteristics of the obstacle can be stably captured, the tracking drift risk caused by the shielding of the obstacle part, the change of the gesture or the change of the scale is reduced, and the continuity and the reliability of the tracking process are ensured. In one embodiment, the performing movement control for the robot based on the obstacle tracking position includes: and controlling the robot to stop moving under the condition that the distance between the robot and the obstacle is determined to meet a first preset distance condition based on the obstacle tracking position. In the above embodiment, in the case that it is determined that the distance between the robot and the obstacle meets the first preset distance condition, the robot is controlled to stop moving, automatic obstacle stopping of the robot is achieved, and the possibility that the robot collides with the obstacle is reduced. In one embodiment, after the controlling the robot to stop moving, the method further comprises: And in a preset time period, performing obstacle tracking based on wide-angle images continuously acquired by the wide-angle camera, and controlling the robot to restart moving under the condition that the distance between the robot and the obstacle is determined to meet a second preset distance condition. In the above embodiment, in the preset duration, the obstacle of the wide-angle image is continuously tracked, and under the condition that the distance between the robot and the obstacle is determined to meet the second preset distance condition, the robot