CN114307100B - Shooting training method and system based on automatic cruise robot - Google Patents
Shooting training method and system based on automatic cruise robot Download PDFInfo
- Publication number
- CN114307100B CN114307100B CN202111606067.6A CN202111606067A CN114307100B CN 114307100 B CN114307100 B CN 114307100B CN 202111606067 A CN202111606067 A CN 202111606067A CN 114307100 B CN114307100 B CN 114307100B
- Authority
- CN
- China
- Prior art keywords
- basketball
- robot
- automatic cruise
- cruise robot
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
- Manipulator (AREA)
Abstract
The invention discloses a shooting training method and system based on an automatic cruise robot, and belongs to the technical field of automatic cruise robots. The method comprises the steps of firstly constructing a high-precision map of a basketball court, then acquiring image information around the automatic cruise robot, respectively tracking a basketball and a player on the court in real time to obtain the positions of the basketball and the player on the court relative to the automatic cruise robot, obtaining the real-time distances between the basketball and the player on the court and between the basketball and the robot, carrying out ball picking, ball passing and moving instructions under various conditions based on preset distance thresholds, and controlling the automatic cruise robot to pick up, pass and move the basketball. The automatic identification of the orientation of the player is realized by the shooting training pitching machine, so that the shooting precision and the shooting efficiency are improved.
Description
Technical Field
The invention belongs to the technical field of automatic cruise robots, and relates to a shooting training method and system based on an automatic cruise robot.
Background
Basketball is one of the most popular ball games worldwide, and people enjoy basketball more and more. The shooting training is the necessary training of each basketball player every day, and in a professional team, a special trainer can finish the ball picking and passing work, or teammates can help each other to pick and pass when training together. However, when one person alone goes to practice shooting, a lot of time and physical strength are wasted due to the need of frequently running the ball picking-up, so that the training effect is greatly reduced. The existing shooting training service robot on the market can only serve the basketball towards the same direction because a net higher than a basket is needed to block the basketball, and the orientation of the athlete can not be automatically identified, so that the sight line, the shooting precision and the shooting efficiency of the athlete can be influenced.
Disclosure of Invention
The invention aims to overcome the defects that in the prior art, the shooting training pitching machine cannot automatically identify the orientation of a player to influence shooting precision and shooting efficiency, and provides a shooting training method and a shooting training system based on an automatic cruise robot.
In order to achieve the purpose, the invention adopts the following technical scheme to realize the purpose:
a shooting training method based on an automatic cruise robot comprises the following steps:
step 1) constructing a high-precision map of a basketball court, and determining the position of an automatic cruise robot in the high-precision map;
step 2) acquiring image information around the automatic cruise robot, simultaneously respectively tracking the basketball and the athletes on the field in real time, respectively acquiring the positions of the basketball and the athletes on the field relative to the automatic cruise robot, and acquiring real-time distances between the basketball and the athletes on the field and between the basketball and the robot;
step 3) controlling the cruise robot to carry out ball picking operation based on the positions of the basketball and the sportsman on the field relative to the automatic cruise robot and the real-time distance between the basketball and the sportsman on the field;
respectively planning paths of the automatic cruise robot towards all basketballs for the basketballs with the distance more than 5 meters away from the athletes, selecting the basketball farthest away from the athletes, and picking up the basketball according to the planned paths;
if the distance between the basketball and the athlete is less than 5 meters, returning to the state of waiting for picking up the basketball in the step 2);
when the distance between the robot and the picked basketball is less than 0.3m, the automatic cruising robot executes a ball picking command;
step 4), when the sportsman needs to pass the ball, sending a ball passing instruction to the cruise robot, and passing the ball to the sportsman by the cruise robot; and after the ball conveying is finished, repeating the operation of the step 3).
Preferably, in step 2), the specific process of real-time tracking is as follows:
step 2.1) acquiring target tracking videos of basketball and athletes on the field, respectively performing search window processing on each frame of image in the target tracking videos, and extracting chromaticity components from the processed images;
step 2.2) carrying out statistics on the color component to establish a color histogram;
step 2.3) obtaining a reverse projection graph corresponding to each chrominance component based on the color histogram;
step 2.4) reselecting a search window, respectively calculating a zero order moment, a first order moment and a mass center in a new search window, determining the size of the new search window by using the zero order moment, moving the center of the new search window to the position of the mass center, recording the moving distance, converging when the moving distance is smaller than a set limit value, reading in the next frame and entering step 2.1) to restart tracking; and stopping real-time tracking when the moving distance is larger than or equal to the set limit value.
Preferably, step 2.1) is specifically:
a frame including an image of a basketball and a player is captured, a circular target window is selected from the image, the basketball is located in the circular target window, a rectangular target window is selected from the image, the player is located in the rectangular target window, and then the circular target window and the rectangular target window are converted to convert the RGB image into the HSV image.
Preferably, step 2.4) is specifically:
selecting a circular search window by taking the center of a circle of the target window, selecting a rectangular search window by taking the intersection point of diagonal lines of the rectangle as the center, respectively calculating a zero order moment, a first order moment and a mass center in the circular search window and the rectangular search window, and determining the size of a new search window corresponding to each zero order moment;
zero order moment:
M 00 ＝∑ x ∑ y I(x,y) (1)
calculating a first step distance:
M 10 ＝∑ x ∑ y xI(x,y)；M 01 ＝∑ x ∑ y yI(x,y) (2)
calculate the centroid of the search window:
x c ＝M 10 /M 00 ；y c ＝M 01 M 00 (3)
wherein, I (x, y) represents the gray value of the pixel point;
the specific process of determining the size of the new search window by using the zeroth order moment is as follows: setting the radius of the new search window as a function having a proportional relation with the zero-order moment, and finding out the pixel point I with the maximum gray value in the search window max And setting the radius of the next search window as:
preferably, in step 2), the specific operation after real-time tracking is:
after real-time tracking, obtaining the positions of a basketball and an athlete relative to the automatic cruise robot on the field, obtaining the linear distances between the automatic cruise robot and the basketball and the athlete respectively, and firstly calculating to obtain the angle between the basketball and the automatic cruise robot and the angle between the athlete and the automatic cruise robot;
respectively calculating the coordinates of the basketball and the athlete relative to the automatic cruise robot according to the linear distance and the angle;
calculating coordinates of the basketball and the athlete relative to the automatic cruise robot to obtain a real-time distance between the basketball and the athlete;
wherein x is 1 ，y 1 As coordinates of the basketball relative to the auto-cruise robot, α 1 Is the angle between the connecting line of the basketball and the robot and the center line of the robot, theta is the camera view angle, N is the total number of single-row pixel points in the RGB image, X 1 Is the abscissa of the center point of the sphere, L 1 Is the linear distance, x, from the center of the basketball to the center of the robot, obtained by a laser radar 2 ，y 2 For the coordinates of the athlete relative to the auto-cruise robot, α 2 The angle between the connecting line of the athlete and the robot and the center line of the robot, X 2 Is the abscissa of the center point of the sphere, L 2 The linear distance between the athlete and the center of the robot is obtained by the laser radar, and L is the linear distance between the basketball and the athlete.
Preferably, in step 3), planning a path of the auto-cruise robot to the basketball on each court, specifically:
acquiring a two-dimensional grid map in an application scene based on the step 1), and mapping the coordinates of the automatic cruise robot and the basketball appearing in the image in the step 2) to the two-dimensional grid map to obtain the position relation of the automatic cruise robot and each basketball in the image;
and planning paths from the automatic cruise robot to the basketballs by adopting an A-star algorithm, calculating straight line distances between the athletes and the basketballs, comparing the distances between the athletes and different basketballs, selecting the basketball farthest from the athletes, and picking up the basketball farthest from the athletes according to the planned paths.
An auto-cruise robot-based basketball shooting training system, comprising:
the camera module is interacted with the control module and is used for acquiring image information around the automatic cruise robot, respectively tracking the basketball and the sportsman in the field in real time and transmitting the acquired image information and the real-time tracked video information to the control module;
the laser radar positioning module is interacted with the control module and is used for constructing a high-precision map of the whole basketball court, determining the position of the automatic cruise robot in the whole map and transmitting information to the control module;
the ball picking module is mutually interacted with the control module and is used for receiving a ball picking command of the control module, further controlling the automatic cruise robot to pick up the basketball and then feeding data after ball picking back to the control module;
the AR identification module is respectively interacted with the camera module and the control module, and is used for judging the intention of the athlete according to the face information of the athlete given by the camera module and transmitting a signal to the control module;
the pass module is interacted with the control module and is used for receiving a pass command given by the control module and further controlling the automatic cruise robot to pass the ball to the athlete;
the control module is used for receiving information fed back by the laser radar positioning module, the camera module, the ball picking module, the AR identification module and the ball passing module, and calculating the positions of the basketball and the player on the field relative to the automatic cruise robot and the real-time distance between the basketball and the player on the field based on the acquired image information and the real-time tracked video information; and planning a path of the automatic cruise robot facing the basketball, controlling the automatic cruise robot to move towards the basketball, sending a ball picking command to the ball picking module and sending a ball passing command to the ball passing module.
Preferably, in the process of controlling the passing, when the AR identification module identifies that the eyes of the athlete are looking at the automatic cruise robot and stay for more than 3 seconds, the control module sends a passing command to the passing module;
when the straight line distance between the automatic cruise robot and the basketball is less than 0.3m, the control module sends a ball picking command to the ball picking module.
Compared with the prior art, the invention has the following beneficial effects:
the invention discloses a shooting training method based on an automatic cruise robot, which comprises the steps of firstly constructing a high-precision map of a basketball court, then obtaining image information around the automatic cruise robot, respectively tracking a basketball and a player on the court in real time to obtain the positions of the basketball and the player on the court relative to the automatic cruise robot, obtaining the real-time distances between the basketball and the player on the court and between the basketball and the robot, and carrying out ball picking, ball passing and moving instructions under various conditions based on a preset distance threshold value to control the automatic cruise robot to pick up, pass and move the basketball. The automatic identification of the orientation of the player is realized by the shooting training pitching machine, so that the shooting precision and the shooting efficiency are improved.
The invention discloses a shooting training system based on an automatic cruise robot, which is characterized in that a high-precision map of the whole basketball court is constructed through a laser radar, and the position of the automatic cruise robot in the whole map is determined; shooting by using a camera to obtain image information around the automatic cruise robot, respectively tracking the basketball and the players on the field in real time, and calculating the positions of the basketball and the players on the field relative to the automatic cruise robot and the real-time distance between the basketball and the players on the field; selecting a basketball with a distance greater than five meters from the player, planning a global path and a local path of the automatic cruise robot towards the basketball, selecting the basketball farthest from the player, and picking up the basketball; when the AR technology identifies that the player intends to want to take the ball to the automatic cruise robot, the ball passing controller receives the ball passing command of the central processor. The automatic cruise robot can accurately position the athlete and the basketball and track the movement track of the basketball in real time; the ball picking and passing functions can be rapidly realized according to the instruction. The manpower cost of the team is reduced, the time and the physical strength of the athletes are saved, and the training efficiency of the athletes is improved.
Drawings
FIG. 1 is a structural diagram of a ball picking and passing system of an automatic cruise robot in an embodiment 1;
fig. 2 is a flow chart of the ball picking and passing method of the automatic cruise robot in the embodiment 2.
Wherein, 1-laser radar positioning module; 2-a camera module; a 3-AR identification module; 4-a control module; 5-ball picking module; 6-pass module.
Detailed Description
The invention is described in further detail below with reference to the accompanying drawings:
example 1
A shooting training method based on an automatic cruise robot comprises the following steps:
step 1) constructing a high-precision map of a basketball court, and determining the position of an automatic cruise robot in the high-precision map;
step 2) acquiring image information around the automatic cruise robot, simultaneously respectively tracking the basketball and the athletes on the field in real time, respectively acquiring the positions of the basketball and the athletes on the field relative to the automatic cruise robot, and acquiring real-time distances between the basketball and the athletes on the field and between the basketball and the robot;
step 3) controlling the cruise robot to carry out ball picking operation based on the positions of the basketball and the sportsman on the field relative to the automatic cruise robot and the real-time distance between the basketball and the sportsman on the field;
respectively planning paths of the automatic cruise robot towards all the basketballs for the basketballs which are more than 5 meters away from the players, selecting the basketball which is farthest away from the players, and picking up the basketball according to the planned paths;
if the distance between the basketball and the player is less than 5 meters, returning to the state of waiting for picking up the basketball in the step 2);
when the distance between the robot and the picked basketball is less than 0.3m, the automatic cruising robot executes a ball picking command;
step 4), when the sportsman needs to pass the ball, sending a ball passing instruction to the cruise robot, and passing the ball to the sportsman by the cruise robot; and after the ball conveying is finished, repeating the operation of the step 3).
Example 2
Referring to fig. 2, the method comprises the steps of:
s101, constructing a high-precision map of the whole basketball court by using a laser radar, and determining the position of the automatic cruise robot in the whole map;
specifically, a high-precision map of the basketball court is constructed by using a laser radar module fixed on the automatic cruise robot, and the position of the automatic cruise robot in the whole map is determined and transmitted to the central processing unit.
And S102, shooting by using a camera to acquire image information around the automatic cruise robot, respectively tracking the basketball and the athletes on the field in real time, calculating the positions of the basketball and the athletes on the field relative to the automatic cruise robot and the real-time distance between the basketball and the athletes on the field, and transmitting the positions to the control module, the AR identification module and the laser radar module.
Specifically, real-time positions of the athlete and the basketball are automatically tracked through a camera module arranged at the top of the head of the automatic cruise robot and are respectively transmitted to a control module, an AR recognition module and a laser radar module.
Specifically, the specific implementation manner in step S102 is as follows:
(1) Acquiring a video shot by a camera, intercepting a first frame comprising images of a basketball and an athlete, manually selecting a round target window to enable the window to just include the whole basketball, selecting a rectangular target window to enable the window to just include the whole athlete, and then respectively converting two target window images from an RGB image into an HSV image;
(2) Extracting a chrominance component H from (1) the HSV image;
(3) Counting the H component extracted in the step (2), establishing a color histogram of the chromaticity H, namely, averagely dividing a color space into N intervals, wherein each interval becomes a subinterval of the histogram, the value corresponding to each subinterval from left to right is (0,1,2...., N-1), and counting the number of pixel points in each subinterval;
(4) The method comprises the steps of converting images of a first frame of a video shot by a camera, including basketball and athletes, into HSV space images;
(5) And (3) acquiring the chromaticity H value of each pixel point in the HSV image, finding out the subinterval where the value is located in the color histogram established in the step (3) according to the chromaticity H value, and replacing the value at the position, which is the same as the HSV space image, in the reverse projection image with the value corresponding to the subinterval, so as to obtain the reverse projection image.
(6) Respectively selecting a circular search window and a rectangular search window by taking the center of a circle of the target window and the intersection point of rectangular diagonal lines as centers, respectively calculating a zero order moment, a first order moment and a mass center in the search windows, and determining the size of a new search window by using the zero order moment; moving the center of the search window to the position of the mass center calculated in the step (6), recording the moving distance a, converging when the moving distance is smaller than a set limit value, reading in the next frame and entering the step (4), or directly entering the step (6);
calculating the zero order moment:
M 00 ＝∑ x ∑ y I(x,y) (1)
calculating a first step distance:
M 10 ＝∑ x ∑ y xI(x,y)；M 01 ＝∑ x ∑ y yI(x,y) (2)
calculate the centroid of the search window:
x c ＝M 10 /M 00 ；y c ＝M 01 M 00 (3)
wherein, I (x, y) represents the gray value of the pixel point.
Setting the radius of the new search window as a function having a proportional relation with the zero-order moment, and finding out the pixel point I with the maximum gray value in the search window max And setting the radius of the next search window as:
specifically, the formula for solving the angles between the basketball and the automatic cruise robot, the player and the automatic cruise robot, the coordinates of the basketball and the player relative to the automatic cruise robot, and the real-time linear distance between the basketball and the player is as follows:
wherein x is 1 ，y 1 Is the coordinate of the basketball relative to the auto-cruise robot, alpha 1 Is the angle between the connecting line of the basketball and the robot and the center line of the robot, theta is the camera view angle, N is the total number of single-line pixel points in the RGB image, X 1 Is the abscissa of the center point of the sphere, L 1 The distance between the center of the basketball and the center of the robot is obtained by the laser radar. x is the number of 2 ，y 2 As coordinates of the athlete relative to the auto-cruise robot, alpha 2 Is the angle between the connecting line of the athlete and the robot and the center line of the robot, X 2 Is the abscissa of the center point of the sphere, L 2 The linear distance between the athlete and the center of the robot is obtained by the laser radar, and L is the linear distance between the basketball and the athlete.
And S103, selecting basketballs with the distance larger than 5 m according to the positions of the automatic cruise robot in the drawing in S101, the positions of the basketballs on the field calculated by the image shot by the camera in S102 relative to the automatic cruise robot and the real-time distance between the basketballs and the players, and planning the path of the automatic cruise robot to each basketball. If no basketball with the distance more than five meters exists, the basketball stops working; if so, selecting the basketball farthest away, and picking up the basketball according to the planned path.
Processing the surrounding environment of the automatic cruise robot and the whole basketball court by adopting a laser radar to obtain a two-dimensional grid map under an application scene, and mapping the coordinates of the automatic cruise robot and the basketball appearing in the image to the two-dimensional grid map to obtain the position relation of the automatic cruise robot and each basketball in the image; and planning paths from the robot to the basketballs by adopting an A-x algorithm, calculating straight line distances between the players and the basketballs, comparing the distances, selecting the basketball farthest from the players, and picking up the basketball according to the planned path.
And S104, the central processor module sends a ball picking command to the ball picking module.
Specifically, when the central processor module calculates that the linear distance between the automatic cruise robot and the basketball is less than 0.3m, a ball picking command is sent to the ball picking module.
And S105, the AR identification module processes the video information transmitted by the camera module and transmits the identification information to the central processor.
Specifically, video information sent by a camera module arranged at the top of the head of the robot is identified by an AR identification module, and when the face of the athlete is identified to be looking at the automatic cruise robot and the time is more than 3 seconds, the information that the athlete has the intention to take a ball is transmitted to a central processor.
And S106, the central processor module sends a pass command to the pass module.
Specifically, when the central processor module receives the information that the player has the intention to play, which is sent by the AR identification module, the central processor module sends a pass command to the ball transfer module.
Example 3
An auto-cruise robot-based basketball shooting training system, comprising:
the camera module is interacted with the control module and is used for acquiring image information around the automatic cruise robot, respectively tracking the basketball and the sportsman in the field in real time and transmitting the acquired image information and the real-time tracked video information to the control module;
the laser radar positioning module is interacted with the control module and used for constructing a high-precision map of the whole basketball court, determining the position of the automatic cruise robot in the whole map and transmitting information to the control module;
the ball picking module is interacted with the control module and is used for receiving a ball picking command of the control module, further controlling the automatic cruise robot to pick up a basketball and then feeding back data after ball picking to the control module;
the AR identification module is respectively interacted with the camera module and the control module, and is used for judging the intention of the athlete according to the face information of the athlete given by the camera module and transmitting a signal to the control module;
the pass module is interacted with the control module and is used for receiving a pass command given by the control module and further controlling the automatic cruise robot to pass the ball to the athlete;
the control module is used for receiving information fed back by the laser radar positioning module, the camera module, the ball picking module, the AR identification module and the ball passing module, and calculating the positions of the basketball and the player on the field relative to the automatic cruise robot and the real-time distance between the basketball and the player on the field based on the acquired image information and the real-time tracked video information; and planning a path of the automatic cruise robot facing the basketball, controlling the automatic cruise robot to move towards the basketball, sending a ball picking command to the ball picking module and sending a ball passing command to the ball passing module.
Example 4
Referring to fig. 1, the system comprises a laser radar positioning module 1, a camera module 2, an AR identification module 3, a control module 4, a ball picking module 5 and a ball passing module 6.
Specifically, the laser radar positioning module 1 is arranged in front of the chest of the automatic cruise robot and used for constructing a high-precision map of the whole basketball court, determining the position of the automatic cruise robot in the whole map, receiving data transmitted by the camera module in real time, calculating the linear distances between an athlete and the automatic cruise robot and between each basketball and the automatic cruise robot and transmitting the linear distances to the control module;
the camera module 2 is arranged at the head of the automatic cruise robot and used for shooting and acquiring image information around the automatic cruise robot, respectively tracking the basketball and the athletes on the field in real time, calculating the positions of the basketball and the athletes on the field relative to the automatic cruise robot and the real-time distance between the basketball and the athletes on the field, and transmitting video information to the laser radar module, the control module and the AR identification module;
the AR recognition module 3 is arranged at the head of the automatic cruise robot and used for judging the intentions of athletes according to the face information of the athletes given by the camera module and transmitting signals to the central processor;
the control module 4 receives the position of the automatic cruise robot in the whole map, the linear distance between each player and the automatic cruise robot, the linear distance between each basketball and the automatic cruise robot, the position is transmitted by the laser radar positioning module 1, the video information of the basketball and the players is received, the video information of the basketball and the video information of the players is transmitted by the camera module 2, the face identification information of the players is received, the ball picking feedback is transmitted by the ball picking module 5, the ball passing feedback is received, the linear distance between the player and the basketball is calculated, a ball picking command is sent to the ball picking module 5, and a ball passing command is sent to the ball passing module 6.
The ball picking module 5 is arranged on the arm of the automatic cruising robot, and controls the automatic cruising robot to pick up the basketball and feed the basketball back to the central processor after receiving a ball picking command of the central processor module;
and the pass module 6 is arranged on the arm of the automatic cruise robot and used for passing balls to the athletes when receiving a pass command given by the central processor.
In conclusion, the shooting training method based on the automatic cruise robot, provided by the invention, realizes automatic tracking of the basketball, and commands the robot to complete ball picking and passing tasks; the labor cost of the team is reduced, the time and the physical strength of the athletes are saved, and the training efficiency of the athletes is improved.
The above-mentioned contents are only for illustrating the technical idea of the present invention, and the protection scope of the present invention is not limited thereby, and any modification made on the basis of the technical idea of the present invention falls within the protection scope of the claims of the present invention.
Claims (7)
1. A shooting training method based on an automatic cruise robot is characterized by comprising the following steps:
step 1) constructing a high-precision map of a basketball court, and determining the position of the automatic cruise robot in the high-precision map;
step 2) acquiring image information around the automatic cruise robot, simultaneously respectively tracking the basketball and the athletes on the field in real time, respectively acquiring the positions of the basketball and the athletes on the field relative to the automatic cruise robot, and acquiring real-time distances between the basketball and the athletes on the field and between the basketball and the robot;
in step 2), the specific process of real-time tracking is as follows:
step 2.1) acquiring target tracking videos of basketball and athletes on the field, respectively performing search window processing on each frame of image in the target tracking videos, and extracting chromaticity components from the processed images;
step 2.2) carrying out statistics on the color component to establish a color histogram;
step 2.3) obtaining a reverse projection graph corresponding to each chrominance component based on the color histogram;
step 2.4) reselecting a search window, respectively calculating a zero order moment, a first order moment and a mass center in a new search window, determining the size of the new search window by using the zero order moment, moving the center of the new search window to the position of the mass center, recording the moving distance, converging when the moving distance is smaller than a set limit value, reading in the next frame and entering step 2.1) to restart tracking; stopping real-time tracking when the moving distance is greater than or equal to a set limit value;
step 3) controlling the cruise robot to carry out ball picking operation based on the positions of the basketball and the sportsman on the field relative to the automatic cruise robot and the real-time distance between the basketball and the sportsman on the field;
respectively planning paths of the automatic cruise robot towards all the basketballs for the basketballs which are more than 5 meters away from the players, selecting the basketball which is farthest away from the players, and picking up the basketball according to the planned paths;
if the distance between the basketball and the athlete is less than 5 meters, returning to the state of waiting for picking up the basketball in the step 2);
when the distance between the robot and the picked basketball is less than 0.3m, the automatic cruising robot executes a ball picking command;
step 4), when the sportsman needs to pass the ball, sending a ball passing instruction to the cruise robot, and passing the ball to the sportsman by the cruise robot; and after the ball conveying is finished, repeating the operation of the step 3).
2. The auto-cruise-robot-based shooting training method according to claim 1, characterized in that step 2.1) is specifically:
a frame including an image of a basketball and a player is captured, a circular target window is selected from the image, the basketball is located in the circular target window, a rectangular target window is selected from the image, the player is located in the rectangular target window, and then the circular target window and the rectangular target window are converted to convert the RGB image into the HSV image.
3. The automatic cruise robot-based shooting training method according to claim 2, characterized in that step 2.4) is specifically:
selecting a circular search window by taking the center of a circle of the target window, selecting a rectangular search window by taking the intersection point of diagonal lines of the rectangle as the center, respectively calculating a zero order moment, a first order moment and a mass center in the circular search window and the rectangular search window, and determining the size of a new search window corresponding to each zero order moment;
zero order moment:
M 00 ＝∑ x ∑ y I(x，y) (1)
calculating a first step distance:
M 10 ＝∑ x ∑ y xI(x，y)；M 01 ＝∑ x ∑ y yI(x，y) (2)
calculate the centroid of the search window:
x c ＝M 10 /M 00 ；y c ＝M 01 M 00 (3)
wherein, I (x, y) represents the gray value of the pixel point;
the specific process of determining the size of the new search window by using the zeroth order moment is as follows: setting the radius of the new search window as a function having a proportional relation with the zero-order moment, and finding out the pixel point I with the maximum gray value in the search window max And setting the radius of the next search window as:
4. the automatic cruise robot-based shooting training method according to claim 1, wherein in step 2), the specific operation after real-time tracking is as follows:
after real-time tracking, the positions of a basketball and a player on the field relative to the automatic cruise robot are obtained, the straight line distances between the automatic cruise robot and the basketball and between the automatic cruise robot and the player are obtained, and firstly, the angle between the basketball and the automatic cruise robot and the angle between the player and the automatic cruise robot are obtained through calculation;
respectively calculating the coordinates of the basketball and the athlete relative to the automatic cruise robot according to the linear distance and the angle;
calculating coordinates of the basketball and the athlete relative to the automatic cruise robot to obtain a real-time distance between the basketball and the athlete;
wherein x is 1 ，y 1 Is the coordinate of the basketball relative to the auto-cruise robot, alpha 1 Is the angle between the connecting line of the basketball and the robot and the center line of the robot, theta is the camera view angle, N is the total number of single-row pixel points in the RGB image, X 1 Is the abscissa of the center point of the sphere, L 1 Is the linear distance, x, from the center of the robot to the center of the basketball obtained by the laser radar 2 ，y 2 As coordinates of the athlete relative to the auto-cruise robot, alpha 2 Is the angle between the connecting line of the athlete and the robot and the center line of the robot, X 2 Is the abscissa of the center point of the sphere, L 2 The linear distance between the player and the center of the robot is obtained by the laser radar, and L is the linear distance between the basketball and the player.
5. The shooting training method based on the automatic cruise robot as claimed in claim 1, wherein in step 3), the path of the automatic cruise robot towards the basketball on each court is planned, specifically:
acquiring a two-dimensional grid map in an application scene based on the step 1), and mapping the coordinates of the automatic cruise robot and the basketball appearing in the image in the step 2) to the two-dimensional grid map to obtain the position relation of the automatic cruise robot and each basketball in the image;
and planning paths from the automatic cruise robot to the basketballs by adopting an A-star algorithm, calculating straight line distances between the athletes and the basketballs, comparing the distances between the athletes and different basketballs, selecting the basketball farthest from the athletes, and picking up the basketball farthest from the athletes according to the planned paths.
6. A basketball shooting training system for an automatic cruise robot capable of implementing the method of claim 1, comprising:
the camera module is interacted with the control module and is used for acquiring image information around the automatic cruise robot, respectively tracking the basketball and the sportsman in the field in real time and transmitting the acquired image information and the real-time tracked video information to the control module;
the laser radar positioning module is interacted with the control module and used for constructing a high-precision map of the whole basketball court, determining the position of the automatic cruise robot in the whole map and transmitting information to the control module;
the ball picking module is interacted with the control module and is used for receiving a ball picking command of the control module, further controlling the automatic cruise robot to pick up a basketball and then feeding back data after ball picking to the control module;
the AR identification module is respectively interacted with the camera module and the control module, and is used for judging the intention of the athlete according to the face information of the athlete given by the camera module and transmitting a signal to the control module;
the ball passing module is interacted with the control module and is used for receiving a ball passing command given by the control module and further controlling the automatic cruise robot to pass balls to the athletes;
the control module is used for receiving information fed back by the laser radar positioning module, the camera module, the ball picking module, the AR identification module and the ball passing module, and calculating the positions of the basketball and the player on the field relative to the automatic cruise robot and the real-time distance between the basketball and the player on the field based on the acquired image information and the real-time tracked video information; and a path of the automatic cruise robot towards the basketball is planned, the automatic cruise robot is controlled to move towards the basketball, and a ball picking command and a ball passing command are sent to the ball picking module and the ball passing module respectively.
7. The auto-cruise-robot-based basketball shooting training system according to claim 6, wherein the control module sends a pass command to the pass module when the AR recognition module recognizes that the eyes of the athlete are looking at the auto-cruise robot and stay for more than 3 seconds in the process of controlling the pass;
when the linear distance between the automatic cruise robot and the basketball is less than 0.3m, the control module sends a ball picking command to the ball picking module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111606067.6A CN114307100B (en) | 2021-12-25 | 2021-12-25 | Shooting training method and system based on automatic cruise robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111606067.6A CN114307100B (en) | 2021-12-25 | 2021-12-25 | Shooting training method and system based on automatic cruise robot |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114307100A CN114307100A (en) | 2022-04-12 |
| CN114307100B true CN114307100B (en) | 2023-04-11 |
Family
ID=81012470
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111606067.6A Active CN114307100B (en) | 2021-12-25 | 2021-12-25 | Shooting training method and system based on automatic cruise robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114307100B (en) |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130178314A1 (en) * | 2012-01-11 | 2013-07-11 | John Michael Raimonda | Tethered ball attached to a strap for batting practice off an existing tee |
| CN103316463B (en) * | 2013-07-11 | 2015-07-22 | 西北农林科技大学 | Rotatable basketball shooting practicing machine |
| CN108673527B (en) * | 2018-06-06 | 2024-05-07 | 昆明理工大学 | Court service robot |
| CN111558210A (en) * | 2019-09-20 | 2020-08-21 | 西北工业大学 | Ball picking method of full-automatic ball picking robot and full-automatic ball picking robot |
-
2021
- 2021-12-25 CN CN202111606067.6A patent/CN114307100B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN114307100A (en) | 2022-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102571076B1 (en) | Device, system, and method for tracking an object using radar data and imager data | |
| US20230092516A1 (en) | Computer-implemented method for automated detection of a moving area of interest in a video stream of field sports with a common object of interest | |
| CN101537618B (en) | Visual system for ball picking robot in stadium | |
| CN106197422B (en) | A kind of unmanned plane positioning and method for tracking target based on two-dimensional tag | |
| AU2018404876B2 (en) | Golf ball tracking system | |
| CN110653819B (en) | System and method for generating kicking action of humanoid robot | |
| CN110929596A (en) | Shooting training system and method based on smart phone and artificial intelligence | |
| CN110533687A (en) | Multiple target three-dimensional track tracking and device | |
| US20190002104A1 (en) | Unmanned aerial vehicle escape system, unmanned aerial vehicle escape method, and program | |
| CN107124606A (en) | The long-range image pickup method of digital video advertisement based on unmanned plane | |
| CN114307100B (en) | Shooting training method and system based on automatic cruise robot | |
| CN112245889A (en) | Method and system for automatically judging tennis ball out of bounds | |
| CN103218826A (en) | Projectile detecting, three-dimensional positioning and trajectory predicting method based on Kinect | |
| CN112766033B (en) | Method for estimating common attention targets of downlinks in scene based on multi-view camera | |
| CN111738093B (en) | Automatic speed measuring method for curling balls based on gradient characteristics | |
| CN112720501A (en) | Mechanical arm positioning billiard ball grabbing method and system based on machine vision technology | |
| WO2021056552A1 (en) | Video processing method and device | |
| CN111784749A (en) | Space positioning and motion analysis system based on binocular vision | |
| CN117576156A (en) | Method and system for judging ball landing | |
| Chen et al. | Check for updates Multi-drone Multi-object Tracking with RGB Cameras Using Spatio-Temporal Cues | |
| Sun et al. | A Novel Robust Color Tag for a Highly Efficient Soccer Robot Vision System | |
| Cui et al. | Game Image Detection and Application Based on Improved YOLOv5 | |
| CN111791238B (en) | Control system and control method for accurate medicine spraying robot | |
| CN111450515B (en) | Automatic identification striking system and method for robot shooting confrontation match | |
| CN114973409A (en) | Goal scoring identification method and system based on court environment and personnel pose |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |