阅读说明：本技术 机器人的自定位方法 (Self-positioning method of robot ) 是由 刘晓程 吴忠华 康望星 周煜申 于 2019-11-15 设计创作，主要内容包括：本发明公开一种机器人的自定位方法,包括：在GPS信号无法进行定位时,用机器人携带的双目摄像头寻找房间内的墙角图像,与墙角库中的墙角进行匹配,获得墙角坐标信息；并依据三个以上的墙角坐标信息进行机器人的自定位。本发明根据双目测距原理确定机器人与墙角之间的距离,分别确定任意三个墙角与机器人之间的距离；然后根绝三边测距原理,已知墙角的坐标和墙角到机器人的距离,计算出机器人在世界坐标系下的三维坐标,即完成机器人的自定位；从而实现了在无GPS情况下对机器人自身的定位。(The invention discloses a self-positioning method of a robot, which comprises the following steps: when the GPS signal can not be used for positioning, a binocular camera carried by the robot is used for searching a corner image in a room, and the corner image is matched with a corner in a corner library to obtain corner coordinate information; and the self-positioning of the robot is carried out according to the coordinate information of more than three corners. The method comprises the steps of determining the distance between the robot and a corner according to a binocular distance measuring principle, and respectively determining the distance between any three corners and the robot; then, according to the trilateral distance measurement principle, knowing the coordinates of the corner and the distance from the corner to the robot, calculating the three-dimensional coordinates of the robot in a world coordinate system, and finishing the self-positioning of the robot; therefore, the robot can be positioned under the condition of no GPS.)

1. A method for self-positioning a robot, the method comprising:

when the GPS signal can not be used for positioning, a binocular camera carried by the robot is used for searching a corner image in a room, and the corner image is matched with a corner in a corner library to obtain corner coordinate information; and the self-positioning of the robot is carried out according to the coordinate information of more than three corners.

2. A method of self-positioning of a robot according to claim 1, characterized in that the method comprises: the GPS signal is detected according to the following steps:

and comparing the detection information fed back by the detection program with the signal intensity required by the standard GPS for accurate positioning, and if the detected GPS signal intensity is less than the signal intensity required by the standard GPS for positioning, determining that the GPS signal cannot be positioned.

3. The self-positioning method of a robot according to claim 1, characterized in that said method is embodied as:

(1) establishing a wall corner set library taking a room as a unit;

(2) when the GPS signal can not be used for positioning, a camera carried by the robot is used for searching a corner image in a room and matching the corner image with a corner in a corner library to obtain corner coordinate information;

(3) the binocular camera that the robot carried measures the actual distance with the corner, accomplishes the self-align of robot through three corners.

4. A method of self-positioning of a robot according to claim 1, characterized in that the method comprises: the method also comprises the step of carrying out preliminary screening of the rooms in a corner database with huge data volume by using RGB or HSI color characteristics.

5. A method of self-positioning of a robot according to claim 1, characterized in that the method comprises: the wall corner set library is established by the following steps:

1) sampling wall corner images according to sampling rules

2) Correcting the sampling image;

3) and extracting the corner feature descriptors.

6. The self-positioning method of a robot of claim 5, wherein said method comprises: and 4) storing the wall angle classification.

7. The self-positioning method of the robot as claimed in claim 3, wherein the binocular camera carried by the robot in the step (3) measures the actual distance from the corner, and the self-positioning of the robot is accomplished through three corners:

determining the distance between the first corner and the robot; determining the distance between the second corner and the robot; and determining the distance between the third corner and the robot; according to the trilateral distance measurement principle, the coordinates of the corner and the distance from the corner to the robot are known, and the three-dimensional coordinates of the robot in a world coordinate system are calculated, so that the self-positioning of the robot is completed.

Technical Field

The invention relates to a robot self-positioning method and system combining a GPS (global positioning system) and a corner.

Background

According to the prediction of experts, the proportion of the aged over 60 years old in China reaches 20% by 2025, wherein the aged over 80 years old exceeds thirty million people, and the need of health care homes for service robots is urgent. Taking an indoor service type mobile robot as an example, in order to efficiently and reliably control the robot, it is necessary to determine the position of the robot itself, i.e., a self-positioning problem. Therefore, sensing and fusing information by using various sensors to complete reliable positioning is the primary function of the autonomous mobile robot.

The most widely used global application is GPS positioning. In three-dimensional space, ideally, only three satellites are needed to achieve GPS positioning. However, GPS positioning relies on satellite signals, which are degraded under building shelters, resulting in poor positioning effect of the indoor mobile robot.

The positioning method combining GPS positioning and wall corner positioning can greatly improve the positioning precision and speed of the robot.

Disclosure of Invention

Aiming at the problems, the invention provides a robot self-positioning method combining a GPS and a corner.

In order to achieve the above object, the present invention provides a self-positioning method of a robot, comprising:

when the GPS signal can not be used for positioning, a binocular camera carried by the robot is used for searching a corner image in a room, and the corner image is matched with a corner in a corner library to obtain corner coordinate information; and the self-positioning of the robot is carried out according to the coordinate information of more than three corners.

Further, the method comprises the following steps: the GPS signal is detected according to the following steps:

and comparing the detection information fed back by the detection program with the signal intensity required by the standard GPS for accurate positioning, and if the detected GPS signal intensity is less than the signal intensity required by the standard GPS for positioning, determining that the GPS signal cannot be positioned.

Further, the method specifically comprises the following steps:

(1) establishing a wall corner set library taking a room as a unit;

(2) when the GPS signal can not be used for positioning, a camera carried by the robot is used for searching a corner image in a room and matching the corner image with a corner in a corner library to obtain corner coordinate information;

(3) the binocular camera that the robot carried measures the actual distance with the corner, accomplishes the self-align of robot through three corners.

Further, the method comprises the following steps: the method also comprises the step of carrying out preliminary screening of the rooms in a corner database with huge data volume by using RGB or HSI color characteristics.

Further, the method comprises the following steps: the wall corner set library is established by the following steps:

1) sampling wall corner images according to sampling rules

2) Correcting the sampling image;

3) extracting wall corner feature descriptors;

further, the method comprises the following steps: and 4) storing the wall angle classification.

Further, the step (3) of measuring the actual distance from the wall corner by using a binocular camera carried by the robot, and the self-positioning of the robot through the three wall corners is specifically as follows:

determining the distance between the first corner and the robot; determining the distance between the second corner and the robot; and determining the distance between the third corner and the robot; according to the trilateral distance measurement principle, the coordinates of the corner and the distance from the corner to the robot are known, and the three-dimensional coordinates of the robot in a world coordinate system are calculated, so that the self-positioning of the robot is completed.

The method comprises the steps of determining the distance between the robot and a corner according to a binocular distance measuring principle, and respectively determining the distance between any three corners and the robot; then, according to the trilateral distance measurement principle, knowing the coordinates of the corner and the distance from the corner to the robot, calculating the three-dimensional coordinates of the robot in a world coordinate system, and finishing the self-positioning of the robot; therefore, the robot can be positioned under the condition of no GPS.

Drawings

Fig. 1 is a flow chart of the robot self-positioning of the present invention.

Fig. 2 is a principle of binocular ranging of the present invention.

Fig. 3 is a diagram of the principles of trilateration of the present invention.

Detailed Description

The invention is further described with reference to the accompanying drawings.