Table tennis playing robot for physical fitness

Abstract

The table tennis playing robot comprises a bottom plate used for being installed at one end of a table tennis table, the upper side of the bottom plate is connected with a moving block through a horizontal moving device, and the upper side of the moving block is fixedly connected with a vertically-arranged first telescopic air cylinder; a top plate is fixedly connected to the upper end of the first telescopic air cylinder, an inclined plate is connected to the upper side of the top plate through an angle inclination device, a third telescopic air cylinder is fixedly connected to the side wall of the inclined plate, and a racket is fixedly connected to the output end of the third telescopic air cylinder. According to the method, the action is flexible, the racket can be controlled to reach any position to complete various complicated ball hitting actions, and the table tennis ball track drop point position can be predicted by constructing the track drop point prediction model, so that the accurate ball hitting action is completed.

Description

Technical Field

[0001] The present invention relates to the technical field of table tennis robots, and particularly to a table tennis rally robot for physical fitness. Background Art

[0002] Existing table tennis robots are mainly divided into two types: one is a robot with a hitting mechanism that can complete simple hitting actions and hit the table tennis ball back. Due to its special structure, the hitting action is slow and the movement flexibility is poor. It cannot effectively hit back a table tennis ball flying at high speed and cannot hit a high-quality table tennis ball, so it has no practical value in actual table tennis practice; the other is to transform an industrial robotic arm robot and install a table tennis racket at the end of the robotic arm. Because of its complex structure, immobility, and high cost, it has no practical promotion value. Summary of the Invention

[0003] The present invention provides a table tennis rally robot for physical fitness to solve the problems raised in the above background art.

[0004] To achieve the above object, the present invention adopts the following technical solutions:

[0005] A table tennis rally robot for physical fitness includes a bottom plate for installation at one end of a table tennis table. A moving block is connected to the upper side of the bottom plate through a horizontal moving device. A vertically arranged first telescopic cylinder is fixedly connected to the upper side of the moving block. The upper end of the first telescopic cylinder is fixedly connected to a top plate. An inclined plate is connected to the upper side of the top plate through an angle tilting device. A third telescopic cylinder is fixedly connected to the side wall of the inclined plate. The output end of the third telescopic cylinder is fixedly connected to a racket.

[0006] It further includes a control mechanism, and the control mechanism includes a detection unit, an analysis unit, and a control unit.

[0007] Preferably, the horizontal moving device includes two first fixing plates fixedly connected to the upper sides of both ends of the bottom plate. A threaded rod is rotatably connected between the two first fixing plates. A motor is fixedly connected to the side wall of one of the first fixing plates. The output shaft of the motor penetrates through the side wall of the first fixing plate and is fixedly connected to one end of the threaded rod. A moving block is threadedly connected to the threaded rod, and the moving block is slidably connected to the upper side of the bottom plate.

[0008] Preferably, the angle tilting device includes two second fixing plates which are fixedly connected to the upper side of the top plate. The bottom end of the tilting plate is rotationally connected between the two second fixing plates through a rotating shaft. A second telescopic cylinder is rotationally connected to the upper side of the top plate. The upper end of the second telescopic cylinder is rotationally connected to a sliding block. Two third fixing plates are fixedly connected to the side wall of the tilting plate. A fixing rod is fixedly connected between the two third fixing plates. The sliding block is slidably connected to the fixing rod.

[0009] Preferably, the detection unit includes a plurality of high-speed cameras installed around the table tennis table, as well as an accelerometer and a gyroscope installed on the table tennis ball.

[0010] The analysis unit is used to capture the flight trajectory of the table tennis ball through the high-speed camera, extract the position of the table tennis ball, and then obtain the accelerometer and gyroscope on the table tennis ball through the wireless transmission module to obtain the moving acceleration and rotational angular velocity of the table tennis ball in real time. Then, the real-time position, moving acceleration, and rotational angular velocity are input into the trajectory landing point prediction model to obtain the trajectory landing point position at the next moment.

[0011] Preferably, the high-speed camera captures the flight trajectory of the table tennis ball and extracts the position of the table tennis ball, specifically as follows:

[0012] Step 1, high-speed image acquisition: Use multiple high-speed cameras with a frame rate ≥ 1000fps to obtain images of the table tennis ball in flight to ensure capturing the details of the high-speed movement of the table tennis ball.

[0013] Step 2, image preprocessing: First, use Gaussian filtering or median filtering to remove the image noise captured by the high-speed camera to improve the target detection accuracy. Then, convert the image from the RGB color space to the HSV color space to facilitate segmentation using the yellow feature of the table tennis ball.

[0014] Step 3, target detection: Through HSV threshold segmentation, set the HSV threshold for the yellow range to generate a binary mask, where the HSV threshold is H: 20 - 30, S: 100 - 255, V: 100 - 255.

[0015] Step 4, target tracking based on motion features: Calculate the pixel difference between adjacent frames of the image, detect the motion area, and achieve the tracking of the table tennis ball.

[0016] Step 5, 3D reconstruction: Calibrate the internal and external parameters of the camera, establish the relationship between the camera coordinate system and the world coordinate system, then match the corresponding points in two adjacent frame images, calculate the disparity map, and then calculate the three-dimensional coordinates of the table tennis ball according to the disparity and camera parameters.

[0017] Step 6, data post-processing: Use moving average or Kalman filtering to smooth the position data and remove the noise.

[0018] Preferably, the specific construction steps of the trajectory landing point prediction model are as follows:

[0019] First step: Obtain a data set composed of a large number of table tennis rally samples, and each sample in the data set includes four indicators: real-time position, moving acceleration, rotational angular velocity, and the trajectory landing point position at the next moment.

[0020] Second step: Use the data set to train a machine learning model. The machine learning model takes the three indicators of real-time position, moving acceleration, and rotational angular velocity after standardization as inputs, and the trajectory landing point position at the next moment as the output, and obtains the trajectory landing point prediction model through supervised learning.

[0021] Preferably, for the target table tennis ball whose trajectory landing point needs to be measured, first measure the three indicators of the real-time position, moving acceleration, and rotational angular velocity of the table tennis ball, standardize them, and input them into the trajectory landing point prediction model to obtain the prediction result of the trajectory landing point position at the next moment.

[0022] Preferably, in the first step, divide the data set into a training set and a validation set, use the training set to train multiple machine learning models simultaneously, and use the validation set to evaluate the performance accuracy of different machine learning models, and select the machine learning model with the best performance to construct the trajectory landing point prediction model.

[0023] Preferably, in the second step, the selected machine learning model is at least one of logistic regression, linear discriminant analysis, K-nearest neighbor, naive Bayes, support vector machine, random forest, and neural network.

[0024] Preferably, the control unit is used to control the racket to reach the corresponding position in advance according to the trajectory landing point position of the table tennis ball obtained by the trajectory landing point prediction model, and control the racket to hit the table tennis ball.

[0025] Compared with the prior art, the beneficial effects of the present invention are:

[0026] The present invention is not only flexible in movement, can control the racket to reach any position, and complete various complex hitting actions, but also can predict the trajectory landing point position of the table tennis ball by constructing a trajectory landing point prediction model, so as to complete accurate hitting actions.

[0027] The above description is only an overview of the technical solution of the present invention. In order to be able to understand the technical means of the present invention more clearly and implement it according to the content of the specification, the following takes the preferred embodiments of the present invention and combines with the accompanying drawings to describe in detail as follows. The specific implementation manners of the present invention are given in detail by the following embodiments and their accompanying drawings. Brief Description of the Drawings

[0028] The accompanying drawings described herein are used to provide a further understanding of the present invention and form a part of this application. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation to the present invention. In the drawings:

[0029] Figure 1 is a schematic diagram of the overall structure of a table tennis rally robot for physical fitness proposed by the present invention;

[0030] Figure 2 is a schematic diagram of the upper side structure of the bottom plate proposed by the present invention;

[0031] Figure 3 is a schematic side view structure of a table tennis rally robot for physical fitness proposed by the present invention;

[0032] Figure 4 is a schematic diagram of the control mechanism structure proposed by the present invention.

[0033] In the drawings, the list of components represented by each reference numeral is as follows:

[0034] 1. Motor; 2. Moving block; 3. Threaded rod; 4. Bottom plate; 5. First fixing plate; 6. First telescopic cylinder; 7. Second fixing plate; 8. Top plate; 9. Second telescopic cylinder; 10. Tilt plate; 11. Racket; 12. Third fixing plate; 13. Fixed rod; 14. Sliding block; 15. Third telescopic cylinder. Detailed implementation manners

[0035] The principles and features of the present invention are described below with reference to the accompanying drawings. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention. In the following paragraphs, the present invention is described more specifically by way of example with reference to the drawings. It should be noted that the drawings are all in a very simplified form and use non-precise scales, only for the purpose of facilitating and clearly assisting in explaining the purpose of the embodiments of the present invention.

[0036] It should be noted that when a component is referred to as being "fixed to" another component, it can be directly on the other component or there can also be an intermediate component. When a component is considered to be "connected to" another component, it can be directly connected to the other component or there may be an intermediate component at the same time. When a component is considered to be "disposed on" another component, it can be directly disposed on the other component or there may be an intermediate component at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are only for the purpose of illustration.

[0037] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which this invention belongs. The terms used in the description of the present invention herein are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0038] Please refer to Figures 1 to 4 , in an embodiment of the present invention, a table tennis rally robot for physical fitness includes a bottom plate 4 for being installed at one end of a table tennis table. A moving block 2 is connected to the upper side of the bottom plate 4 through a horizontal moving device. The horizontal moving device includes two first fixing plates 5. The two first fixing plates 5 are fixedly connected to the upper sides of both ends of the bottom plate 4. A threaded rod 3 is rotatably connected between the two first fixing plates 5. A motor 1 is fixedly connected to the side wall of one of the first fixing plates 5. The output shaft of the motor 1 penetrates through the side wall of the first fixing plate 5 and is fixedly connected to one end of the threaded rod 3. The moving block 2 is threadedly connected to the threaded rod 3, and the moving block 2 is slidably connected to the upper side of the bottom plate 4;

[0039] A vertically arranged first telescopic cylinder 6 is fixedly connected to the upper side of the moving block 2. The upper end of the first telescopic cylinder 6 is fixedly connected to a top plate 8. The upper side of the top plate 8 is connected to an inclined plate 10 through an angle tilting device. The angle tilting device includes two second fixing plates 7. The two second fixing plates 7 are fixedly connected to the upper side of the top plate 8. The bottom end of the inclined plate 10 is rotatably connected between the two second fixing plates 7 through a rotating shaft. A second telescopic cylinder 9 is rotatably connected to the upper side of the top plate 8. The upper end of the second telescopic cylinder 9 is rotatably connected to a sliding block 14. Two third fixing plates 12 are fixedly connected to the side wall of the inclined plate 10. A fixing rod 13 is fixedly connected between the two third fixing plates 12. The sliding block 14 is slidably connected to the fixing rod 13. A third telescopic cylinder 15 is fixedly connected to the side wall of the inclined plate 10. The output end of the third telescopic cylinder 15 is fixedly connected to a racket 11;

[0040] It further includes a control mechanism, and the control mechanism includes a detection unit, an analysis unit, and a control unit;

[0041] The detection unit includes a plurality of high-speed cameras installed around the table tennis table, as well as an accelerometer and a gyroscope installed on the table tennis ball;

[0042] The analysis unit is used to capture the flight trajectory of the table tennis ball through the high-speed camera, extract the position of the table tennis ball, and then obtain the accelerometer and gyroscope on the table tennis ball through a wireless transmission module to obtain the moving acceleration and rotational angular velocity of the table tennis ball in real time. Then, the real-time position, moving acceleration, and rotational angular velocity are input into the trajectory landing point prediction model to obtain the trajectory landing point position at the next moment.

[0043] Among them, a high-speed camera captures the flight trajectory of the table tennis ball and extracts the position of the table tennis ball. Specifically:

[0044] Step 1, high-speed image acquisition: Use multiple high-speed cameras with a frame rate ≥ 1000fps to obtain images of the table tennis ball in flight, ensuring that the details of the high-speed movement of the table tennis ball are captured;

[0045] Step 2, image preprocessing: First, use Gaussian filtering or median filtering to remove the image noise captured by the high-speed camera and improve the target detection accuracy. Then, convert the image from RGB to the HSV color space to facilitate segmentation using the yellow feature of the table tennis ball;

[0046] Step 3, target detection: Through HSV threshold segmentation, set the HSV threshold for the yellow range to generate a binary mask, where the HSV threshold is H: 20 - 30, S: 100 - 255, V: 100 - 255;

[0047] Step 4, target tracking based on motion features: Calculate the pixel difference between adjacent frames of the image, detect the motion area, and achieve the tracking of the table tennis ball;

[0048] Step 5, 3D reconstruction: Calibrate the internal and external parameters of the camera, establish the relationship between the camera coordinate system and the world coordinate system, then match the corresponding points in two adjacent frame images, calculate the disparity map, and then calculate the 3D coordinates of the table tennis ball according to the disparity and camera parameters;

[0049] Step 6, data post-processing: Use moving average or Kalman filtering to smooth the position data and remove noise.

[0050] The specific construction steps of the trajectory landing point prediction model are as follows:

[0051] The first step is to obtain a dataset composed of a large number of table tennis rally samples, and each sample in the dataset includes four indicators: real-time position, moving acceleration, rotational angular velocity, and the trajectory landing point position at the next moment. Among them, the dataset is divided into a training set and a validation set. Use the training set to train multiple machine learning models simultaneously, and use the validation set to evaluate the performance accuracy of different machine learning models, and select the machine learning model with the best performance to construct the trajectory landing point prediction model;

[0052] The second step is to train the machine learning model using the dataset. Among them, the machine learning model takes the three indicators of real-time position, moving acceleration, and rotational angular velocity after standardization as inputs, and the trajectory landing point position at the next moment as the output, and obtains the trajectory landing point prediction model through supervised learning. Among them, the selected machine learning model is at least one of logistic regression, linear discriminant analysis, K-nearest neighbor, naive Bayes, support vector machine, random forest, and neural network.

[0053] For a target table tennis ball whose trajectory landing point needs to be measured, first measure three indicators of the real-time position, moving acceleration, and rotational angular velocity of the table tennis ball. After standardizing them, input them into the trajectory landing point prediction model to obtain the prediction result of the trajectory landing point position at the next moment.

[0054] The control unit is used to control the table tennis ball trajectory landing point position obtained by the trajectory landing point prediction model. The control unit can control the forward and reverse rotation of 3 through 1. The forward and reverse rotation of the threaded rod 3 can drive the moving block 2 to move horizontally left and right, thereby driving the racket 11 to move left and right. The control unit can drive the racket 11 to move up and down through the first telescopic cylinder 6, and thus can pre-control the racket 11 to reach the corresponding position of the table tennis ball trajectory landing point. After that, by controlling the expansion and contraction of the second telescopic cylinder 9, the pitch angle of the inclined plate 10 can be changed, thereby changing the hitting angle of the racket 11. Finally, after the table tennis ball approaches the racket 11, controlling the third telescopic cylinder 15 to extend can control the racket 11 to hit the table tennis ball.

[0055] The above is only the preferred embodiment of the present invention, and does not impose any form of limitation on the present invention; any ordinary technician in the industry can smoothly implement the present invention as shown in the accompanying drawings of the specification and as described above; however, any equivalent changes made by those skilled in the art within the scope of the technical solution of the present invention by using the technical content disclosed above for minor modifications, decorations, and evolutions are equivalent embodiments of the present invention; at the same time, any equivalent changes, modifications, and evolutions made to the above embodiments based on the essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A table tennis rally robot for physical fitness, comprising a bottom plate (4) for installation at one end of a table tennis table, characterized in that, On the upper side of the bottom plate (4), a moving block (2) is connected through a horizontal moving device. On the upper side of the moving block (2), a vertically arranged first telescopic cylinder (6) is fixedly connected. At the upper end of the first telescopic cylinder (6), a top plate (8) is fixedly connected. On the upper side of the top plate (8), an inclined plate (10) is connected through an angle tilting device. On the side wall of the inclined plate (10), a third telescopic cylinder (15) is fixedly connected. At the output end of the third telescopic cylinder (15), a racket (11) is fixedly connected. It further includes a control mechanism, and the control mechanism includes a detection unit, an analysis unit, and a control unit.

2. The table tennis rally robot for physical fitness according to claim 1, wherein The horizontal moving device includes two first fixing plates (5). The two first fixing plates (5) are fixedly connected to the upper sides of both ends of the bottom plate (4). A threaded rod (3) is rotatably connected between the two first fixing plates (5). On the side wall of one of the first fixing plates (5), a motor (1) is fixedly connected. The output shaft of the motor (1) penetrates through the side wall of the first fixing plate (5) and is fixedly connected to one end of the threaded rod (3). The moving block (2) is threadedly connected to the threaded rod (3), and the moving block (2) is slidably connected to the upper side of the bottom plate (4).

3. A table tennis rally robot for physical fitness according to claim 2, characterized in that, The angle tilting device includes two second fixing plates (7). The two second fixing plates (7) are fixedly connected to the upper side of the top plate (8). The bottom end of the inclined plate (10) is rotatably connected between the two second fixing plates (7) through a rotating shaft. On the upper side of the top plate (8), a second telescopic cylinder (9) is rotatably connected. The upper end of the second telescopic cylinder (9) is rotatably connected to a sliding block (14). On the side wall of the inclined plate (10), two third fixing plates (12) are fixedly connected. A fixed rod (13) is fixedly connected between the two third fixing plates (12). The sliding block (14) is slidably connected to the fixed rod (13).

4. A table tennis rally robot for physical fitness according to claim 3, characterized in that, The detection unit includes a plurality of high-speed cameras installed around the table tennis table, as well as an accelerometer and a gyroscope installed on the table tennis ball. The analysis unit is used to capture the flight trajectory of the table tennis ball through the high-speed camera, extract the position of the table tennis ball, and then obtain the accelerometer and gyroscope on the table tennis ball through the wireless transmission module to obtain the moving acceleration and rotational angular velocity of the table tennis ball in real time. Then, the real-time position, moving acceleration, and rotational angular velocity are input into the trajectory landing point prediction model to obtain the trajectory landing point position at the next moment.

5. A table tennis rally robot for physical fitness according to claim 4, characterized in that, The high-speed camera captures the flight trajectory of the table tennis ball and extracts the position of the table tennis ball, specifically as follows: Step 1, high-speed image acquisition: Use multiple high-speed cameras with a frame rate ≥ 1000fps to obtain images of the table tennis ball in flight to ensure capturing the details of the high-speed movement of the table tennis ball. Step 2, image preprocessing: First, use Gaussian filtering or median filtering to remove the image noise captured by the high-speed camera to improve the target detection accuracy. Then, convert the image from RGB to the HSV color space to facilitate segmentation using the yellow feature of the table tennis ball. Step 3, Object Detection: Through HSV threshold segmentation, set the HSV threshold for the yellow range to generate a binary mask, where the HSV threshold is H: 20 - 30, S: 100 - 255, V: 100 - 255; Step 4, Object Tracking Based on Motion Features: Calculate the pixel differences between adjacent frames of the image, detect the moving area, and achieve the tracking of the table tennis ball; Step 5, 3D Reconstruction: Calibrate the internal and external parameters of the camera, establish the relationship between the camera coordinate system and the world coordinate system, then match the corresponding points in two adjacent frame images, calculate the disparity map, and then calculate the 3D coordinates of the table tennis ball according to the disparity and camera parameters; Step 6, Data Post - processing: Use moving average or Kalman filter to smooth the position data and remove noise.

6. The table tennis rally robot for physical fitness according to claim 5, characterized in that, The specific construction steps of the trajectory landing point prediction model are as follows: The first step: Obtain a dataset composed of a large number of table tennis rally samples, and each sample in the dataset includes four indicators: real - time position, moving acceleration, rotational angular velocity, and the trajectory landing point position at the next moment; The second step: Use the dataset to train a machine learning model. The machine learning model takes the three indicators of real - time position, moving acceleration, and rotational angular velocity after standardization as inputs, and the trajectory landing point position at the next moment as the output, and obtains the trajectory landing point prediction model through supervised learning.

7. A table tennis rally robot for physical fitness according to claim 6, characterized in that, For the target table tennis ball whose trajectory landing point needs to be measured, first measure the three indicators of the real - time position, moving acceleration, and rotational angular velocity of the table tennis ball, standardize them, and then input them into the trajectory landing point prediction model to obtain the prediction result of the trajectory landing point position at the next moment.

8. A table tennis rally robot for physical fitness according to claim 7, characterized in that, In the first step, divide the dataset into a training set and a validation set, use the training set to train multiple machine learning models simultaneously, and use the validation set to evaluate the performance accuracy of different machine learning models, and select the machine learning model with the best performance to construct the trajectory landing point prediction model.

9. The table tennis rally robot for physical fitness according to claim 8, characterized in that, In the second step, the selected machine learning model is at least one of logistic regression, linear discriminant analysis, K - nearest neighbor, naive Bayes, support vector machine, random forest, and neural network.

10. A table tennis rally robot for physical fitness according to claim 9, characterized in that, The control unit is used to control the racket (11) to reach the corresponding position in advance according to the trajectory landing point position of the table tennis ball obtained by the trajectory landing point prediction model, and control the racket (11) to hit the table tennis ball.

Images