CN109079788B

CN109079788B - Chess playing method based on humanoid robot and humanoid robot

Info

Publication number: CN109079788B
Application number: CN201810961056.1A
Authority: CN
Inventors: 庄礼鸿; 王文豪
Original assignee: Xiamen University of Technology
Current assignee: Xiamen University of Technology
Priority date: 2018-08-22
Filing date: 2018-08-22
Publication date: 2020-04-14
Anticipated expiration: 2038-08-22
Also published as: CN109079788A

Abstract

The invention discloses a chess playing method based on a humanoid robot and the humanoid robot. Wherein the method comprises the following steps: humanoid robot carries out image processing to the image of the chess board that this camera obtained, obtain the pixel position of chess board angular point, and then the pixel position according to this chess board angular point that obtains, set up a target location, and then according to the target location of this setting, adopt monocular distance measurement calculation mode, calculate the actual position who obtains this chess board angular point, and then the actual position according to this chess board angular point that this calculation obtained, adopt the accurate control of kinematics to snatch the chess piece from the chess board and play this target location to the chess board. Through the mode, the humanoid robot can automatically grab the chess pieces from the chess board to play to the specified position of the chess board.

Description

Chess playing method based on humanoid robot and humanoid robot

Technical Field

The invention relates to the technical field of humanoid robots, in particular to a chess playing method based on a humanoid robot and the humanoid robot.

Background

In 2007, a hot tide of research of humanoid robot NAO (an artificial intelligence robot developed by Aldebaran Robotics corporation) was raised at home and abroad, and the appearance of the humanoid robot NAO replaces a Sony robot dog to become a standard platform.

In the humanoid robot field, some colleges in europe and america are in the front of research, and many other scholars learn their algorithms and creative thinking one after another. The study of the trainees is various, some of the trainees pay attention to the study of algorithms such as image recognition, positioning algorithm, voice processing and the like, and other trainees are interested in studying the motion bionics of humanoid robots, wherein many of the studies are for competition, and others are for further improving the research work.

In China, the research on the field of anthropomorphic robots starts late, and only less than 5 colleges or research units in China use anthropomorphic robots in the research process until 2010, and the results are few. Relatively speaking, the university of science and technology in china, the university of sienna traffic, the university of coordination, etc., have advanced in this respect, wherein schuman et al, the university of science and technology in china, has achieved a modest result, and mainly studies the motion-related engine of humanoid robots.

Although the research on the humanoid robot field in China starts late, the development speed is very rapid, and until 2013, the related scientific research work of humanoid robots is carried out in most colleges and universities of '985' and '211'. Meanwhile, many competitions about humanoid robots are held in China for further communication and study. In addition, in the field of image processing and positioning algorithms associated with humanoid robots, scholars in our country have also achieved a number of results, including image processing algorithms and positioning algorithms in colleges and universities. Therefore, although the research on the field of humanoid robots in China started late, scholars and researchers are still learning and exploring humanoid robots. However, the existing humanoid robot research scheme can not realize that the humanoid robot can automatically grab the chess pieces from the chess board and play the chess pieces to the specified positions of the chess board.

Disclosure of Invention

In view of the above, the present invention provides a chess playing method based on a humanoid robot and the humanoid robot, which can realize that the humanoid robot can automatically grab a chess piece from a chess board to play to a specified position of the chess board.

According to one aspect of the invention, a humanoid robot-based chess playing method is provided, which comprises the following steps:

the humanoid robot carries out image processing on the image of the chess board acquired by the camera to obtain the pixel position of the corner point of the chess board;

setting a target position according to the obtained pixel position of the corner point of the chessboard;

calculating to obtain the actual position of the corner point of the chessboard of the chess by adopting a monocular distance measurement calculation mode according to the set target position;

and according to the actual positions of the angular points of the chessboard of the chess, which are obtained by calculation, capturing chess pieces from the chessboard and playing the chess pieces to the target positions of the chessboard by adopting kinematics accurate control.

Wherein, humanoid robot carries out image processing to the image of the chess board that obtains, obtains the pixel position of chess board angular point, includes:

the humanoid robot adopts the open source computer vision storehouse mode, carries out image processing to the image of chess board that obtains the pixel position of chess board angular point.

Wherein, humanoid robot adopts open source computer vision storehouse mode, carries out image processing to the image of the chess board that obtains, obtains the pixel position of chess board angular point, includes:

humanoid robot adopts open source computer vision storehouse mode, carries out the image processing of grey scale conversion to the image of the chess board that obtains, right the image processing that the harris angular point detected is carried out to the image after grey scale conversion, and is right the image processing that the image binarization was carried out to the image after the harris angular point detected, and according to the image after the image binarization processing obtains the pixel position of chess board angular point.

Wherein, humanoid robot basis adopt monocular distance measurement calculation mode according to the target location who sets up, calculate and obtain the actual position of chess chessboard angular point includes:

the humanoid robot is according to the target location that sets up marks the camera, obtains the inside and outside parameter of camera, according to the model of monocular distance measurement calculation mode is established to the inside and outside parameter of the camera that obtains, and according to the model of monocular distance measurement calculation mode is established to the inside and outside parameter of the camera that obtains, calculates and obtains the actual position of chess chessboard angular point.

Wherein, humanoid robot basis calculate and obtain the actual position of chess board angular point adopts the accurate control of kinematics to snatch the chess piece from the chess board and play to the target position of chess board, include:

the humanoid robot is according to calculate obtaining the actual position of chess chessboard angular point adopts the accurate control right hand level of kinematics to unfold, then the control right hand grabs the piece of getting chess, then the control lifts up right hand arm and moves to the chess chessboard target position.

Wherein humanoid robot carries out image processing to the image of the chess board that the camera obtained, before the pixel position that obtains chess board angular point, still includes:

the humanoid robot obtains the image of the chess board in a camera shooting mode.

According to another aspect of the present invention, there is provided an anthropomorphic robot comprising:

the chess playing control system comprises an image processing module, a position setting module, an angular point position processing module and a chess playing control module;

the image processing module is used for carrying out image processing on the image of the chess board acquired by the camera to obtain the pixel position of the corner point of the chess board;

the position setting module is used for setting a target position according to the obtained pixel position of the corner point of the chessboard of the chess;

the angular point position processing module is used for calculating the actual position of the angular point of the chessboard of the chess by adopting a monocular distance measurement calculation mode according to the set target position;

and the chess playing control module is used for grabbing chess pieces from the chess board to play chess to the target position of the chess board by adopting kinematics accurate control according to the actual positions of the corner points of the chess board obtained by calculation.

The image processing module is specifically configured to:

adopt open source computer vision base mode, carry out the image processing of grey scale conversion to the image of the chess board that obtains, right the image processing that the harris angular point detected is carried out to the image after grey scale conversion, right the image after harris angular point detects carries out the image processing of image binarization, and according to image after the image binarization processing obtains the pixel position of chess board angular point.

The corner position processing module is specifically configured to:

according to the set target position, the camera is calibrated to obtain the internal and external parameters of the camera, a model of a monocular distance measurement calculation mode is established according to the obtained internal and external parameters of the camera, the model of the monocular distance measurement calculation mode is established according to the obtained internal and external parameters of the camera, and the actual position of the chessboard angular point of the chess board is calculated.

Wherein, anthropomorphic robot still includes:

and the camera is used for acquiring the image of the chess board in a shooting mode.

Can discover, above scheme, humanoid robot can adopt the accurate control of kinematics to snatch the chess piece from the chess chessboard and play this target position to the chess chessboard according to the actual position of this chess chessboard angular point that this calculation obtained, can realize humanoid robot and can snatch the assigned position that the chess piece played to the chess chessboard from the chess chessboard automatically.

Furthermore, according to the scheme, the humanoid robot can calibrate the camera according to the set target position to obtain the internal and external parameters of the camera, a model of a monocular distance measurement calculation mode is established according to the obtained internal and external parameters of the camera, the model of the monocular distance measurement calculation mode is established according to the obtained internal and external parameters of the camera, the actual position of the corner point of the chess board is obtained through calculation, and the actual position of the corner point of the chess board with higher accuracy can be obtained through calculation.

Further, above scheme, humanoid robot can adopt the accurate control of kinematics to snatch this target position that chess piece was played to the chess chessboard from the chess chessboard, can realize humanoid robot and pass through the accurate control chess playing operation of chess of kinematics.

Further, above scheme, because two or more humanoid robots of playing chess all the structure the same basically, consequently only need control good time interval, it can to avoid two or more humanoid robots simultaneous motion to lead to the collision, basic thinking is the same with the operation that single humanoid robot carried out chess playing, can realize two or more humanoid robots carry out the operation of playing chess, two or more humanoid robot playing chess system can be defined as two or more humanoid robots accomplish the system of appointed chess step according to the precedence under common operating environment.

Further, above scheme, humanoid robot can acquire the image of chess board through the camera mode of making a video recording, can realize the acquirement to the image of chess board

Drawings

FIG. 1 is a schematic flow chart of an embodiment of the chess playing method based on the humanoid robot of the present invention;

FIG. 2 is an exemplary illustration of an image processing process of a humanoid robot performing Harris corner detection on a gray-level-converted image in an open-source computer vision manner in an embodiment of the method for playing chess based on the humanoid robot;

FIG. 3 is an exemplary illustration of an image processing process of the humanoid robot performing image binarization on an image after Harris corner detection in an open source computer vision manner in an embodiment of the humanoid robot-based chess playing method of the present invention;

FIG. 4 is a schematic diagram of the principle of monocular distance measuring algorithm in one embodiment of the humanoid robot-based chess playing method of the present invention;

FIG. 5 is a schematic diagram illustrating a pixel coordinate system in an embodiment of the method for playing chess based on a humanoid robot according to the present invention;

FIG. 6 is an exemplary illustration of the relationship between the image coordinate system and the pixel coordinate system in an embodiment of the chess playing method based on the humanoid robot;

FIG. 7 is a schematic diagram illustrating a camera coordinate system in an embodiment of the method for playing chess based on a humanoid robot;

FIG. 8 is a schematic diagram illustrating a camera calibration process according to an embodiment of the chess playing method based on a humanoid robot;

FIG. 9 is a schematic diagram illustrating an exemplary compensation algorithm of a monocular distance measuring algorithm in an embodiment of the humanoid robot-based chess playing method of the present invention;

FIG. 10 is another exemplary diagram of a compensation algorithm of a monocular distance measuring algorithm in an embodiment of the humanoid robot-based chess playing method of the present invention;

FIG. 11 is a schematic diagram of an embodiment of a chessboard of the chess according to the method for playing chess based on humanoid robot of the present invention;

FIG. 12 is a schematic view of a humanoid robot performing chess playing operation of chess in one embodiment of the humanoid robot-based chess playing method of the present invention;

FIG. 13 is a schematic flow chart of another embodiment of the method for playing chess based on humanoid robots of the present invention;

FIG. 14 is a schematic structural view of an embodiment of the present invention;

FIG. 15 is a schematic structural view of another embodiment of the present invention;

fig. 16 is a schematic structural view of still another embodiment of the present invention.

Detailed Description

The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be noted that the following examples are only illustrative of the present invention, and do not limit the scope of the present invention. Similarly, the following examples are only some but not all examples of the present invention, and all other examples obtained by those skilled in the art without any inventive work are within the scope of the present invention.

The invention provides a chess playing method based on a humanoid robot, which can realize that the humanoid robot can automatically grab chess pieces from a chess board to play chess to the specified position of the chess board.

Referring to fig. 1, fig. 1 is a schematic flow chart of an embodiment of a method for playing chess based on a humanoid robot according to the present invention. It should be noted that the method of the present invention is not limited to the flow sequence shown in fig. 1 if the results are substantially the same. As shown in fig. 1, the method comprises the steps of:

s101: and the humanoid robot carries out image processing on the image of the chess board acquired by the camera to obtain the pixel position of the corner point of the chess board.

Wherein, carry out image processing at the image of this humanoid robot to the chess board that the camera obtained, before the pixel position that obtains chess board angular point, can also include:

Wherein, this humanoid robot carries out image processing to the image of the chess board that obtains, obtains the pixel position of chess board angular point, can include:

the humanoid robot adopts an OpenCV (Open Source Computer Vision Library) mode to perform image processing on the acquired images of the chess board to obtain the pixel positions of the chess board corners.

Wherein, humanoid robot adopts open source computer vision storehouse OpenCV mode, carries out image processing to the image of the chess board that obtains, obtains the pixel position of chess board angular point, can include:

the humanoid robot adopts an open source computer vision library OpenCV mode to perform image processing of gray level conversion on the acquired image of the chess board, performs image processing of Harris (Harris) angular point detection on the image after gray level conversion, performs image binarization image processing on the image after the Harris angular point detection, and obtains the pixel position of the chess board angular point according to the image after the image binarization processing.

In this embodiment, the humanoid robot adopts an open source computer vision library OpenCV method to perform image processing of gray level conversion on the acquired image of the chess board, and may include:

the open source computer vision library OpenCV's IMREAD (one of computer languages is used for reading a data function in an image file) function supports various dynamic and static image file formats, the file formats supported by different systems are different, but all support BMP (image file Format), and typically also support PNG (Portable Network Graphics ), JPEG (Joint Photographic Experts Group), TIFF (Tag image file Format), etc., this embodiment is an image JPG Format as an example, and then the image is assembled into BGR (Blue-Green-Red, Blue-Green-Red three-color) Format using cv2. cvtctcrocolor (color space conversion function) function, each pixel is represented by a ternary array, and each Integer vector represents a Blue, Green, and Red channel, and other color spaces such as Hue (HSV) channel, Saturation, Value, color model) also represent pixels in the same way, except that the Value range and the number of channels are different, for example, the range of the hue values of the HSV color space is 0-180.

It should be noted that, what the humanoid robot provides by itself is YUV422 (a color coding method) image format, this color space is not common, YUV is a color coding method used by european television, Y has a value in the range of digital value 0-255, indicating brightness, U has a value in the range of digital value 0-255, indicating chroma, V has a value in the range of digital value 0-255, indicating density, Y is separated from other two in YUV color space, the sampling rate of chroma is lower than the luminance sampling rate, and it has an advantage that the image quality does not significantly degrade, and the prototype of YUV is from RGB (Red-Green-Blue ) model, and can be converted to each other by formula.

Referring to fig. 2, fig. 2 is an exemplary view of an image processing process of the humanoid robot performing harris corner detection on the gray-level-converted image by using an open-source computer vision method in an embodiment of the method for playing chess based on the humanoid robot. As shown in fig. 2, the image processing of harris corner detection on the image after the gray conversion by the humanoid robot using an open-source computer vision method may include:

the feature point detection is widely applied to target matching, target tracking, three-dimensional reconstruction and the like, when a target is modeled, target features are extracted from the image after gray level conversion, and common functions comprise color, angular points, feature points, contours, textures and the like; harris corner detection is the basis of feature point detection, and the concept of applying the gray level difference value of adjacent pixels is proposed to determine whether the pixel is a corner, an edge or a smooth area; the principle of Harris corner detection is to calculate the change of gray value in an image by using a moving window; the key process comprises the steps of converting the image into a gray image, calculating a difference image, performing Gaussian smoothing, calculating a local extreme value, determining an angular point and the like; the principle of the corner point is derived from human perception of the corner point, that is, the gray value of each direction of the image is obviously changed; the core of the algorithm is to move the image using a local window to determine that a large change in gray level has occurred; thus, the window is used to calculate the gray scale change of the image, for example: [ -1, 0, 1; -1, 0, 1; -1, 0, 1], [ -1, -1, -1; 0, 0, 0; 1, 1, 1 ]; moving the widget of this function in all directions; as shown in fig. 2, the gray level of the area in the window is greatly changed, and it is considered that the inside of the window meets the corner point, the gray level of the image in the window is not changed, and there is no corner in the window; if the window is moved in a certain direction, the gray level of the image in the window varies greatly, and there is no change in some directions, the image in the window may be a straight line segment.

Referring to fig. 3, fig. 3 is an exemplary view of an image processing process of the humanoid robot performing image binarization on an image detected by harris corner points in an open-source computer vision manner according to an embodiment of the method for playing chess based on the humanoid robot. As shown in fig. 3, the humanoid robot adopts an open-source computer vision mode to perform image binarization image processing on the image after the harris corner detection, so as to obtain pixel positions of the corners of the chessboard of the chess, which may include:

the binarization of the image is to set the gray value of a point on the image to be 0 or 255, i.e. the whole image presents a clear black and white effect, that is, a 256-brightness level gray image is selected by an appropriate threshold value to obtain a two-dimensional image, which can still reflect the whole image and local features of the image; in digital image processing, where binary images occupy a very important position, particularly in actual image processing, there are many systems consisting of binary image processing, the processing and analysis of binary images should be performed first; binarizing the grayscale image to obtain a binarized image, such that, when the image is further processed, the aggregate attribute of the image is only related to the position of pixels having a pixel value of 0 or 255; the pixel is no longer involved, the level values make the processing simple, the data processing and compression small, and in order to obtain a perfect binary image, the non-overlapping regions are often defined using closed and connected boundaries; all pixels whose gray levels are greater than or equal to the threshold value are determined to belong to the specific object and have a gray value of 255; otherwise, these pixels are excluded from the object region and have a gray value of 0, representing the background or special object region; if one object has uniform gray values inside and uniform background with other gray values, a threshold method can be used to obtain a comparative segmentation effect; if the difference between the object and the background is not reflected in the gray scale values, such as different textures, the difference feature can be converted into a gray scale difference, then the image is segmented by using a threshold selection technology, and the threshold is dynamically adjusted to realize that the binary image can dynamically observe the specific result of the segmented image; therefore, the pixel position of the corner point of the chess board can be obtained.

S102: and the humanoid robot sets a target position according to the obtained pixel position of the angular point of the chessboard of the chess.

S103: and the humanoid robot calculates the actual position of the corner point of the chessboard of the chess by adopting a monocular distance measurement calculation mode according to the set target position.

Wherein, this humanoid robot adopts the monocular distance measurement calculation mode according to the target location that should set up, calculates the actual position that obtains this chess chessboard angular point, can include:

the humanoid robot calibrates the camera according to the set target position to obtain internal and external parameters of the camera, establishes a model of a monocular distance measurement calculation mode according to the obtained internal and external parameters of the camera, establishes a model of the monocular distance measurement calculation mode according to the obtained internal and external parameters of the camera, and calculates to obtain the actual position of the chessboard angular point of the chess board.

In the embodiment, the binocular distance measurement algorithm is that two cameras with repeated visual angles shoot within a period of time, and due to different spatial positions of the two cameras, the obtained images can be corrected and transformed, designated targets in the two images are identified, corresponding parameters are respectively obtained, and finally the actual distance between the humanoid robot and the target position is calculated through mathematical derivation; the error ratio of binocular range finding algorithm is less, but the calculation degree of difficulty is bigger, and to humanoid robot, because the coincidence area between two cameras does not hardly, leads to the binocular range finding degree of difficulty can be very big. Therefore, the monocular distance measuring algorithm is preferably adopted in the present embodiment.

Referring to fig. 4, fig. 4 is a schematic diagram illustrating a monocular distance measuring algorithm in an embodiment of the humanoid robot-based chess playing method of the present invention. As shown in fig. 4, x and y are relative position information of the camera, and f is focus information of the camera, in a simple way, the intermediate distance Z is obtained through two similar triangles, but an error caused by the intermediate distance Z is found to be larger in actual calculation, and at this time, a distance measurement model is established to obtain a more accurate distance measurement result; in the embodiment, the internal parameters and the external parameters of the camera are obtained by calibrating the Zhang-friend algorithm of the camera of the humanoid robot, and the conversion between the two-dimensional space image and the three-dimensional space image is obtained in this way.

In this embodiment, the essence of calibrating the camera of the humanoid robot is to establish a relationship between a three-dimensional coordinate image and two-dimensional image coordinates, and if such a relationship can be established, three-dimensional information can be obtained from the two-dimensional image, wherein the coordinate system includes a pixel coordinate system, an image coordinate system, a world coordinate system, and a camera coordinate system, and the description and explanation of the relevant coordinate system will be given below.

Referring to fig. 5, fig. 5 is a schematic diagram illustrating a pixel coordinate system in an embodiment of the humanoid robot-based chess playing method of the present invention. As shown in FIG. 5, the origin of the pixel coordinate system u-v is O0, the abscissa u and the ordinate v are the row and the column of the image respectively, in the open source computer vision library OpenCV, u corresponds to x, v corresponds to y, and the unit of the abscissa and the ordinate is the image pixel.

Referring to fig. 6, fig. 6 is a schematic diagram illustrating a relationship between an image coordinate system and a pixel coordinate system in an embodiment of the method for playing chess based on a humanoid robot according to the present invention. As shown in FIG. 6, the origin of the image coordinate system x-y is O1, which is the midpoint of the pixel coordinate system, assuming (u0, v0) represents the coordinates of O1 in the u-v coordinate system, and dx and dy represent the physical dimensions of each pixel on the horizontal axis x and the vertical axis y, respectively.

Referring to fig. 7, fig. 7 is an exemplary view of a camera coordinate system in an embodiment of the humanoid robot-based chess playing method of the present invention. As shown in fig. 7, O is the optical center of the camera, Zc is the optical axis of the camera, and is perpendicular to the image plane, and OO1 is the focal length of the camera.

In this embodiment, a world coordinate system may also be introduced, where the world coordinate system is introduced for describing the position of the camera, the rotation of any dimension may be expressed as a product of a coordinate vector and a suitable square matrix, the translation vector is an offset of a first coordinate origin and a second coordinate origin, and there are three important parameters in the world coordinate system: the camera comprises a rotation matrix R, a translation vector T and a projection matrix M, wherein the rotation matrix R and the translation vector T are only related to the internal structure of the camera, so the parameters are called as the internal parameters of the camera; the projection matrix M has no relation with the internal structure of the camera itself, so the projection matrix M is an external parameter of the camera, and the process of obtaining the internal parameter and the external parameter is called the calibration of the camera.

Referring to fig. 8, fig. 8 is an exemplary view illustrating a camera calibration performed in an embodiment of the chess playing method based on a humanoid robot according to the present invention. As shown in fig. 8, some photographs of the calibration plates in different directions are obtained by adjusting different shooting angles of the camera, and then corresponding angular points are obtained from the images by applying a relevant feature point function, so as to obtain internal parameters and external parameters of the camera. As shown in fig. 8, the matrix can be solved by a minimum of 6 known points, and in general calibration, several tens of known points can be obtained on the calibration plate, the number of unknowns is much smaller than the number of equations, and in this embodiment, the least square method is used to reduce the error.

After the relative position of the calibrated chessboard of the chess and the humanoid robot is adjusted, a camera of the humanoid robot is used for shooting the calibrated chessboard of the chess, images at different angles are obtained through shooting, then images with preset number, such as 10 images, are selected as calibration images, and the similarity of the images is small; after the calibration mode of the camera is adopted, the internal parameters and the external parameters of the camera are successfully acquired.

In this embodiment, since there may be a relatively large error in the acquisition of the focal length f of the camera in actual operation, secondly, the main optical axis of the camera is not perfectly horizontal, and for convenience, another similar calculation manner of compensating for monocular distance measurement is preferably selected in this embodiment.

Referring to fig. 9 and 10, fig. 9 is a schematic view illustrating a compensation algorithm of a monocular distance measuring algorithm in an embodiment of the humanoid robot-based chess playing method of the present invention, and fig. 10 is another schematic view illustrating a compensation algorithm of a monocular distance measuring algorithm in an embodiment of the humanoid robot-based chess playing method of the present invention. As shown in fig. 9 and 10, the oblique dotted line is the main optical axis of the camera, the included angle between the oblique dotted line and the horizontal ground is b, the boundary of the viewing angle of the camera includes an area that is included by two dotted lines other than the oblique dotted line, the size of the area is 47.64 °, H is the height of the humanoid robot, the approximate pixel coordinates (x, y) of the target position can be obtained through image recognition, and the total size of the image is 640 × 480, in this way, the proportion of the angles in the image and the proportion of the lengths in the image are approximately equal, that is, the size of the angle of a can be obtained, the pitch degree of the humanoid robot head can be obtained, and the distance between the target position and the robot trunk can be obtained through the above method.

S104: the humanoid robot adopts the accurate control of kinematics to snatch the chess piece from the chess board and play to this target position of chess board according to this actual position of calculating this chess board angular point that obtains.

Wherein, this humanoid robot adopts the accurate control of kinematics to snatch the chess piece from the chess chessboard and play this target position to the chess chessboard according to the actual position of this chess chessboard angular point that this calculation obtained, can include:

the humanoid robot adopts the accurate control right hand level of kinematics to unfold according to this chess chessboard angular point's that this calculation obtained actual position, and the piece of chess is grabbed to control the right hand then, and the control is lifted right hand arm and is played this target position of chess chessboard after that.

In the present embodiment, a Webots (a simulation platform of humanoid robots) simulation platform of humanoid robots allows a simulated humanoid robot moving in a virtual world, and Webots can provide a safe and convenient simulation platform for simulating a virtual humanoid robot before testing a real humanoid robot.

In this embodiment, Webots version 7 is a specific version dedicated to simulating humanoid robots, in its platform capable of providing many efficient controllers and simulation experiments of humanoid robots defined in advance, including:

the first step is as follows: starting the simulation robot, selecting a file > to open the file and selecting a file nao.wbt in [ Webots directory ] \ projects \ robots \ nao \ works;

the second step is that: connecting the chord graph to the anthropomorphic robot, starting the chord graph and selecting to connect or click a connection button;

thirdly, testing the behaviors of the humanoid robot in webots;

firstly, ensuring that the computer is successfully connected to the humanoid robot, and then importing the written program to enable the humanoid robot to execute corresponding actions, wherein the humanoid robot needs to ensure that the rigidity is opened, otherwise, action failure can be caused.

In this embodiment, the process of connecting to a real humanoid robot through Webots includes:

the first step is as follows: opening the chord graph;

the second step is that: a connection to a fixed port number 9559 and an IP (Internet Protocol, Protocol for interconnection between networks) address connection is, for example, 192.168.10.106;

the third step: and (5) closing the autonomous mode, constructing a stand up module and controlling the humanoid robot to stand.

Referring to fig. 11, fig. 11 is a schematic view illustrating a chessboard of the chess according to an embodiment of the method for playing chess based on humanoid robots of the present invention. As shown in fig. 11, the humanoid robot may first call the camera to obtain a chessboard picture of the chess, then obtain pixel positions of the angular points, and then obtain positions of actual angular points through a monocular distance measurement algorithm, and as the size of each chess lattice is fixed, the actual positions of the centers of all lattices can be obtained as long as the positions of 1 angular point are obtained.

In this embodiment, this humanoid robot adopts the accurate control of kinematics to snatch the chess piece from the chess board and play this target location to the chess board according to the actual position of this chess board angular point that this calculation obtained, can include:

controlling the right hand of the humanoid robot to horizontally stretch, and ensuring that the right hand arm of the humanoid robot is not touched by a chessboard of the chess to influence the motion of the humanoid robot;

the second step is that: controlling the right hand of the humanoid robot to grab the chess pieces, wherein the arc value at the finger joints of the right hand needs to be carefully adjusted, which is caused by the irregularity of the chess, otherwise the grabbing failure can be caused;

the third step: controlling the humanoid robot to lift the right arm, wherein the lifting height needs to be controlled, because if the lifting height is too high, the chess pieces can fall off, and the chess playing process fails;

the fourth step: controlling the humanoid robot to perform a chess-playing action, wherein which position to play can be adjusted as desired, however, due to the fact that the size of the chess pieces is substantially equal to the size of the grid of the chessboard, combined with slight errors in the algorithm, a small fraction of the chess pieces may fall at other positions, but a large part of the area of the chess pieces can fall at the designated position.

In this embodiment, a design that the double humanoid robots perform chess playing operation of chess may be further provided, and the chess playing system of the double humanoid robots may be defined as a system in which two humanoid robots complete designated chess playing steps in sequence under a common operation environment.

Referring to fig. 12, fig. 12 is a schematic view illustrating a double humanoid robot performing a chess playing operation of chess in an embodiment of the humanoid robot-based chess playing method of the present invention. As shown in fig. 12, since the two humanoid robots playing chess have basically the same structure, only a good time interval needs to be controlled to avoid collision caused by simultaneous movement of the two humanoid robots, and the basic idea is the same as the operation of the single humanoid robot to execute chess playing.

It can be found that in this embodiment, humanoid robot can adopt the accurate control of kinematics to snatch the chess piece from the chess board and play this target location to the chess board according to the actual position of this chess board angular point that this calculation obtained, can realize humanoid robot and can snatch the assigned position that the chess piece played to the chess board automatically from the chess board.

Furthermore, in this embodiment, the humanoid robot can calibrate the camera according to the set target position, obtain the internal and external parameters of the camera, establish the model of the monocular distance measurement calculation mode according to the internal and external parameters of the obtained camera, calculate the actual position of the corner point of the chess board, and can realize calculating the actual position of the corner point of the chess board with higher accuracy.

Further, in this embodiment, humanoid robot can adopt the accurate control of kinematics to snatch this target position that chess piece was played to chess chessboard from the chess chessboard, can realize that humanoid robot passes through the accurate control chess playing operation of chess of kinematics.

Further, in this embodiment, because the structures of two or more humanoid robots playing chess are all basically the same, therefore only need control good time interval, avoid two or more humanoid robots simultaneous motion to lead to the collision can, basic thinking is the same with the operation that single humanoid robot carries out chess playing, can realize that two or more humanoid robots carry out the operation of playing chess, two or more humanoid robot playing chess system can be defined as the system that two or more humanoid robots accomplish appointed playing chess step according to the precedence order under common operating environment.

Referring to fig. 13, fig. 13 is a schematic flow chart of another embodiment of the chess playing method based on humanoid robots of the present invention. In this embodiment, the method includes the steps of:

s1301: the humanoid robot obtains the image of the chess board in a camera shooting mode.

S1302: and the humanoid robot carries out image processing on the image of the chess board acquired by the camera to obtain the pixel position of the corner point of the chess board.

As described above in S101, further description is omitted here.

S1303: and the humanoid robot sets a target position according to the obtained pixel position of the angular point of the chessboard of the chess.

S1304: and the humanoid robot calculates the actual position of the corner point of the chessboard of the chess by adopting a monocular distance measurement calculation mode according to the set target position.

As described above in S103, which is not described herein.

S1305: the humanoid robot adopts the accurate control of kinematics to snatch the chess piece from the chess board and play to this target position of chess board according to this actual position of calculating this chess board angular point that obtains.

As described above in S103, which is not described herein.

It can be found that in this embodiment, humanoid robot can acquire the image of chess board through the camera mode of making a video recording, can realize the acquirement to the image of chess board.

The invention also provides a humanoid robot which can automatically grab the chess pieces from the chess board to play to the specified positions of the chess board.

Referring to fig. 14, fig. 14 is a schematic structural diagram of an embodiment of the present invention. In this embodiment, the humanoid robot 140 is the humanoid robot in the above embodiment, and the humanoid robot 140 includes an image processing module 141, a position setting module 142, a corner point position processing module 143, and a chess playing control module 144.

The image processing module 141 is configured to perform image processing on the image of the chess board acquired by the camera to obtain pixel positions of corner points of the chess board.

The position setting module 142 is configured to set a target position according to the obtained pixel position of the corner point of the chessboard of the chess.

The corner point position processing module 143 is configured to calculate an actual position of the corner point of the chessboard of the chess by using a monocular distance measurement calculation method according to the set target position.

The playing control module 144 is configured to capture chess pieces from the chessboard and play the chess pieces to the target position of the chessboard by using kinematics to accurately control according to the actual positions of the corner points of the chessboard obtained by the calculation.

Optionally, the image processing module 141 may be specifically configured to:

and performing image processing on the acquired image of the chess board by adopting an open source computer vision library mode to obtain the pixel position of the corner point of the chess board.

Optionally, the image processing module 141 may be specifically configured to:

the method comprises the steps of performing image processing of gray level conversion on an acquired image of the chessboard by adopting an open source computer vision library mode, performing image processing of Harris corner detection on the image subjected to gray level conversion, performing image binarization image processing on the image subjected to the Harris corner detection, and obtaining pixel positions of corner points of the chessboard according to the image subjected to the image binarization image processing.

Optionally, the corner position processing module 143 may be specifically configured to:

and calibrating the camera according to the set target position to obtain internal and external parameters of the camera, establishing a model of a monocular distance measurement calculation mode according to the obtained internal and external parameters of the camera, establishing a model of the monocular distance measurement calculation mode according to the obtained internal and external parameters of the camera, and calculating to obtain the actual position of the chessboard corner point of the chess.

Optionally, the chess playing control module 144 may be specifically configured to:

according to the actual position of the corner point of the chessboard of the chess, the right hand is accurately controlled to horizontally stretch by kinematics, then the right hand is controlled to grab the chess pieces of the chess, and then the right hand is controlled to lift up the arm to play the chess to the target position of the chessboard of the chess.

Referring to fig. 15, fig. 15 is a schematic structural view of another embodiment of the present invention. Unlike the previous embodiment, the humanoid robot 150 of the present embodiment further includes: and a camera 151.

The camera 151 is used for acquiring an image of the chess board in a shooting mode.

Each unit module of the humanoid robot 140/150 can respectively execute the corresponding steps in the above method embodiments, so that the detailed description of each unit module is omitted here, and please refer to the description of the corresponding steps above.

Referring to fig. 16, fig. 16 is a schematic structural view of another embodiment of the present invention. The respective unit modules of the humanoid robot can respectively execute the corresponding steps in the above-mentioned method embodiments. For a detailed description of the above method, please refer to the above method, which is not repeated herein.

In this embodiment, the humanoid robot includes: a processor 161, a memory 162 coupled to the processor 161, a setter 163, and a controller 164.

The processor 161 is configured to obtain an image of the chess board by a camera.

The processor 161 is further configured to perform image processing on the acquired image of the chess board to obtain pixel positions of corner points of the chess board.

The memory 162 is used for storing an operating system, instructions executed by the processor 161, and the like.

The setter 163 is configured to set a target position according to the obtained pixel position of the corner point of the chessboard.

The processor 161 is further configured to calculate an actual position of the corner point of the chessboard of the chess by using a monocular distance measurement calculation method according to the set target position.

The controller 164 is configured to capture and play the chess pieces from the chessboard to the target position of the chessboard by using kinematics to accurately control according to the actual positions of the corner points of the chessboard obtained by the calculation.

Optionally, the processor 161 may be specifically configured to:

Optionally, the setter 163 may be specifically configured to:

Optionally, the controller 164 may be specifically configured to:

Further, above scheme, humanoid robot can acquire the image of chess board through the camera mode of making a video recording, can realize the acquireing to the image of chess board.

In the several embodiments provided in the present invention, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, a division of a module or a unit is merely a logical division, and an actual implementation may have another division, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

Units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.

The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be substantially or partially implemented in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, a network device, or the like) or a processor (processor) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

The above description is only a part of the embodiments of the present invention, and not intended to limit the scope of the present invention, and all equivalent devices or equivalent processes performed by the present invention through the contents of the specification and the drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.

Claims

1. A chess playing method based on humanoid robot is characterized by comprising the following steps:

the humanoid robot carries out image processing on the image of the chess board acquired by the camera to obtain the pixel position of the corner point of the chess board; specifically, the humanoid robot performs image processing of gray level conversion on an acquired image of the chess board in an open-source computer vision library mode, performs image processing of Harris corner detection on the image subjected to gray level conversion, performs image binarization image processing on the image subjected to the Harris corner detection, and obtains pixel positions of chess board corners according to the image subjected to the image binarization processing;

2. A method as claimed in claim 1, wherein said humanoid robot calculates the actual position of said angular point of chess board by monocular distance measurement according to said set target position, comprising:

3. A humanoid robot-based chess playing method according to claim 1, wherein said humanoid robot uses kinematics to accurately control grabbing of chess pieces from a chess board to play to said target positions of the chess board based on said calculated actual positions of said chess board corner points, comprising:

4. A humanoid robot-based chess playing method according to any one of claims 1 to 3, characterized in that, before said humanoid robot image-processes the chess board image obtained by the camera to obtain the pixel positions of the chess board corner points, it further comprises:

5. An anthropomorphic robot, comprising:

the image processing module is used for carrying out image processing on the image of the chess board acquired by the camera to obtain the pixel position of the corner point of the chess board; wherein the image processing module is specifically configured to:

performing image processing of gray level conversion on the acquired image of the chess board by adopting an open source computer vision library mode, performing image processing of Harris corner detection on the image subjected to gray level conversion, performing image binarization image processing on the image subjected to the Harris corner detection, and obtaining pixel positions of the chess board corners according to the image subjected to the image binarization processing;

6. The humanoid robot of claim 5, wherein the corner point position processing module is specifically configured to:

7. The humanoid robot as claimed in any one of claims 5 to 6, further comprising: