Body-possessed intelligent tennis training robot and working method thereof

Abstract

The present application relates to a kind of body tennis training robot and its working method;Robot includes wheel foot module, perception module, receiving and sending ball module, feedback module and electric control module;Receiving and sending ball module can control the ball launch pitch angle and the ball speed of robot in the ball launch mode;Switch to the ball receiving mode, can recycle tennis;Feedback module is visualized by light band and display screen Feedback trainer's hitting action;Electric control module includes main control board, analysis board and power component, is responsible for motion control and perception feedback data processing;Main control board coordinates the work of receiving and sending ball module and wheel foot module;Analysis board receives and processes the data from perception module, carries out man-machine interaction with trainer by feedback module;The present application saves the human cost of tennis partner training;Solve the problem that traditional artificial coach is difficult to achieve all-around feeding ball;Make up the deficiency of existing action analysis method and training robot intelligent, significantly improve training efficiency.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of robots, and relates to a body-possessed intelligent tennis training robot and a working method thereof, in particular to a tennis training robot integrating the functions of tennis serving and receiving, and a method of realizing intelligent human-machine interaction by means of body-possessed intelligence. BACKGROUND

[0002] Tennis is favored by a large number of sports enthusiasts due to its competitiveness, interest, sociality and safety, and has a wide range of participants. However, tennis stroke techniques involve complex and delicate movements, and participants are bound to make mistakes in learning and practice, which makes the threshold of tennis advancement high, the learning cost large, and the training period long. In order to master correct tennis movements, a trainer needs to form accurate motor memory through multi-ball practice, and a coach needs to guide and adjust the movements of the trainer by means of objective data analysis means to ensure that the trainer can return the ball at high speed and high accuracy. In order to provide a large number of stable balls, an intelligent serving machine is used for feeding balls; in order to reduce the tedious task of picking up balls, a manual or automatic ball picking tool is usually provided. Therefore, modern tennis training is a complex interactive process involving people, machines and balls.

[0003] However, the current tennis training equipment has the problems of single function, low integration and weak human-machine interaction. The existing training equipment can only complete simple tasks such as fixed-point serving and automatic recovery of tennis balls according to the targets set by players or coaches in advance, lacks real-time perception of the performance of players, and is difficult to generate a scientific and effective guidance scheme in combination with the practice performance of players. In the existing tennis training, the understanding and learning between man and machine are unidirectional. Only by truly improving the mutual understanding and cognition between players and auxiliary equipment, and by giving the training equipment the ability to guide the actions of players, can the training equipment dynamically develop the best training scheme according to the practice performance of players, shorten the learning period of tennis and consolidate correct movements.

[0004] With the rapid development of today's technology, especially the rapid progress of a series of frontier technologies such as artificial intelligence, machine learning and sensor technology. Body-possessed intelligence, compared to traditional artificial intelligence, pays more attention to obtaining information through real-time interaction with the environment, and makes decisions and actions based on these information. By combining body-possessed intelligence method with tennis training robot, the training robot is expected to make greater breakthroughs in perception analysis and interactive decision-making, improve the scientific nature and efficiency of tennis training, and benefit a large number of players. SUMMARY

[0005] The present application proposes a method of driving a tennis training robot to work efficiently by using body-possessed intelligence on the basis of designing a tennis training robot with basic interactive functions. The present application adopts the following technical solutions, which are described in detail in combination with the accompanying drawings.

[0006] The present application provides a tennis training robot, which not only has the basic functions of launching tennis balls in all directions in space and automatically picking up scattered tennis balls in the whole court, but also further integrates the sensing and analysis capabilities of the training player's hitting action posture and the trajectory of the tennis ball. Through visual feedback, the robot can help the training player understand and correct the wrong action more scientifically and efficiently, and ultimately provide an advanced training aid for the training player.

[0007] The embodied intelligent tennis training robot comprises a wheel-foot module 1 and a sensing module 3, characterized in that it further comprises a ball receiving and launching module 2, a feedback module 4 and an electric control module 5.

[0008] The ball receiving and launching module is located above the wheel-foot module, serves as a core component, is responsible for performing the tasks of ball receiving and launching, and can be freely switched between two modes; in the launching mode, the robot's launching pitch angle and launching speed can be controlled; when switched to the ball receiving mode, the tennis ball can be recovered.

[0009] The sensing module is arranged at the top of the robot, is equipped with a binocular camera and an event camera, and is used to capture the action posture of the training player, the position of the robot in the court and the landing point information of the tennis ball.

[0010] The feedback module is arranged on the front of the robot, and the training player's hitting action can be visually fed back through the lamp strip and the display screen.

[0011] The electric control module is arranged inside the robot, comprises a main control board, an analysis board and a power supply and other components, provides hardware support for the motion drive, visual data analysis and control logic of the robot, is responsible for motion control and sensing feedback data processing; the main control board coordinates the work of the ball receiving and launching module and the wheel-foot module; the analysis board receives and processes the data from the sensing module, generates scientific decisions, and performs human-computer interaction with the training player through the feedback module.

[0012] The wheel-foot module mainly comprises a pair of motor wheels, a pair of shanks, a front leg, a front shank, a thigh and four joint motors. When the motor wheels rotate, the robot can move at different speeds and directions; the four joint motors are responsible for driving the flexion and extension actions of the thigh and shank, thereby realizing the height adjustment function of the ball receiving and launching module.

[0013] Further, the ball receiving and launching module mainly comprises a friction wheel mechanism, a ball launching mechanism, a ball storage mechanism and a ball receiving and launching switching mechanism, which are specifically described as follows:

[0014] The friction wheel mechanism drives the friction wheel to rotate at different speeds and directions by a pair of AC brushless motors to achieve the outward launching (serving mode) and inward recovery (receiving mode) of the tennis ball. The pitch adjustment motor controls the angle of the friction wheel fixed support to achieve the angle adjustment of the launching and the control of the angle of the friction wheel during the switching between the serving and receiving modes;

[0015] The serving mechanism includes a three-way serving cylinder, which can push the tennis ball in the serving cylinder to contact with the friction wheel and launch under the action of the electric push rod. When the push rod retreats, the next tennis ball to be launched falls automatically under the action of gravity and enters the launching preparation state.

[0016] The ball storage mechanism is located at the top of the robot, and the received tennis ball falls into the ball storage bin.

[0017] The serving and receiving switching mechanism gives the friction wheel the ability to move forward and backward and swing, which makes the overall mechanism design compact while ensuring the serving and receiving functions of the robot. In the serving mode, the friction wheel mechanism is stored in the abdomen of the robot; in the receiving mode, the friction wheel mechanism is driven by the push rod and the rudder to extend from the abdomen of the robot and swing to the height close to the ground to achieve the pickup of the ground tennis ball.

[0018] Further, under the action of the stirring arm inside the ball storage bin, the bottom tennis ball keeps slow movement, thereby effectively preventing the tennis ball from blocking;

[0019] Further, the right swing arm rudder 2-4-1 of the serving and receiving switching mechanism 2-4 is connected to the right swing arm 2-1-5 in the friction wheel mechanism 2-1 through a bolt, and the left swing arm rudder 2-4-2 is connected to the left swing arm 2-1-6 in the friction wheel mechanism 2-1 through a bolt; under the drive of the rudder, the friction wheel mechanism 2-1 can rotate around the rudder shaft; the right swing arm rudder 2-4-1 is fixed to the right rudder support 2-4-5 through a bolt, and the right rudder support 2-4-5 is fixed to the sliding block of the right guide rail 2-4-3 through a bolt to realize left and right movement; similarly, the left swing arm rudder 2-4-2 is fixed to the left rudder support 2-4-6 through a bolt, and the left rudder support 2-4-6 is fixed to the sliding block of the left guide rail 2-4-4 through a bolt; the two ends of the connecting block 2-4-7 are connected to the right rudder support 2-4-5 and the left rudder support 2-4-6 through bolts, and the middle part of the connecting block 2-4-7 is fixed to the output end of the electric push rod 2-4-8 through a bolt; the rear end of the electric push rod 2-4-8 is fixed to the bottom plate 1-11 through a bolt.

[0020] Further, the wheel-foot module is located at the bottom of the robot, adopts a double-wheel-foot mechanism, and realizes the movement and height adjustment functions of the robot.

[0021] The perception module 3 is composed of a front-facing binocular camera and a rear-facing event camera, which are used to capture the action posture of the trainer, the position of the robot in the court, and the landing point information of the tennis ball.

[0022] The binocular camera in the perception module adopts an existing human body skeleton key point recognition algorithm, which can realize accurate recognition of the position and posture of the trainer. Combined with a target detection algorithm, the binocular camera can calculate the distance between the camera and the boundary of the court, thereby determining the position coordinates of the robot in the court coordinate system. When the binocular camera detects that the trainer performs a swing and hits the ball, the event camera is triggered, which can collect the changing pixel points in the scene at a high speed, and accurately recognize the spatial position coordinates of the tennis ball by analyzing the parallax of the data of the two event cameras.

[0023] The feedback module 4 includes a display screen 4-1 located at the top of the front of the robot and lamp strips distributed on the entire robot body, which are leg lamp strips 4-2, chest lamp strips 4-3, large arm lamp strips 4-4 and small arm lamp strips 4-5 respectively; the arm servo 4-6 is used to control the stretching and tightening of the arm 4-7.

[0024] The feedback module is located on the front of the robot and includes arms, lamp strips and a display screen. The lamp strips are installed on the surface of the double feet, the body and the arms to provide visual feedback for the coordination of the hitting action. The display screen is used to display the geometric posture of the hitting action.

[0025] The present application proposes a somatic intelligent tennis training robot working mode, which includes a serving training mode (referred to as a serving mode) and an autonomous tennis ball recovery mode (referred to as a ball recovery mode).

[0026] 1. Serving mode

[0027] 1) Robot-human-robot space-time position calculation. Establish a court coordinate system; the binocular camera installed in front of the robot captures the position of the court boundary line in real time, and calculates the distance between the camera and the two endpoints of the baseline to determine the real-time position of the robot in the court; the binocular camera captures the position coordinates of the trainer relative to the camera and converts them into the court coordinate system; the binocular event camera installed at the rear of the robot captures the landing position and speed of the tennis ball hit by the trainer.

[0028] 2) The robot serves according to the set rules. The joint motor in the robot wheel module is responsible for adjusting the ball position, and the motor wheel controls the forward and backward movement and the turning angle of the robot; in the ball receiving and sending module, the ball storage mechanism releases the tennis ball to the shooting barrel, the pitch adjusting mechanism controls the pitch angle of the friction wheel, and the rotation speed of the two friction wheels determines the speed and rotation of the ball, the electric push rod in the shooting mechanism pushes the tennis ball to contact the friction wheel, so as to shoot the tennis ball according to the predetermined parameters, realizing the basic ball feeding function.

[0029] 3) The robot analyzes the posture of the trainer and visualizes the feedback to the trainer. The binocular camera captures the spatial position information of the human skeleton in real time, defines the key posture features of the human ball hitting action, and calculates the deviation value of the trainer's posture from the standard features in real time. The robot feeds back the deviation to the trainer in the form of numbers and visual icons on the display screen, reminding the trainer to correct the action.

[0030] 4) The robot analyzes the action coordination of the trainer and visualizes the feedback to the trainer. The binocular camera captures the spatial position and angular velocity information of the hips, chest, shoulders and elbows in the human skeleton in real time; calculates the speed of each part contributing to the racket, compares with the professional player database, evaluates the force degree and timing of each part; through the color change and flow speed of the light belt on the corresponding parts of the robot, the force situation and timing deviation of each part are intuitively displayed, reminding the trainer to correct the action.

[0031] 2. Ball collection mode

[0032] 1) When there is no tennis ball in the ball storage bin, automatically enter the ball collection mode. The robot moves around the court autonomously for one circle, identifies and records the position coordinates of each scattered tennis ball through the binocular camera, and records the serial number of each ball in order.

[0033] 2) After the receiving and sending switching mechanism is started, the wheel-foot module controls the robot to squat to the lowest height. The electric push rod and the rudder work together to move the friction wheel forward and downward, accurately control the speed and angle of ball collection. The robot moves to the back of each tennis ball in order, adjusts its posture to make the center of the friction wheel mechanism align with the center of the tennis ball, and throws the tennis ball into the ball storage bin through the friction wheel mechanism.

[0034] 3) After all the balls with serial numbers are collected, the robot automatically returns to the vicinity of the trainer and waits for instructions to enter the next round of training.

[0035] A working method of a body-integrated intelligent tennis training robot, characterized in that:

[0036] The electronic control module is responsible for motion control and perception feedback data processing; wherein the main control board coordinates the work of the ball receiving and sending module and the wheel-foot module; the analysis board receives and processes data from the perception module, and generates scientific decisions, and interacts with the trainer through the feedback module;

[0037] The wheel-foot module realizes the height adjustment, movement and direction control of the whole robot; four leg motors cooperatively control the bending degree of large and small legs to adjust the ball launching position and ensure the squatting posture when collecting balls; the speed difference of left and right motor wheels determines the turning angle of the robot; when the speed of left and right motor wheels is the same, the robot moves straight forward; when there is a speed difference, the robot turns;

[0038] The ball receiving and sending module has two working states of a serving mode and a ball receiving mode. In the serving mode, the tennis balls in the ball storage mechanism are released into the ball shooting cylinder, the electric push rod pushes the tennis balls to contact with the friction wheels, the friction wheels generate a preset speed and rotation under the driving of the AC brushless motor, the rudder controls the pitch angle, the wheel-foot module controls the serving height and direction, and finally the tennis balls are thrown to the trainer according to the predetermined parameters. In the ball receiving mode, the ball receiving and sending switching mechanism is started, the electric push rod and the rudder work together to move the friction wheel mechanism to the front and lower part of the robot, the ball receiving speed is controlled, the pitch adjusting mechanism adjusts the ball receiving angle, the wheel-foot module controls the robot to squat and adjusts the direction, and finally the tennis balls scattered on the court are collected into the ball storage mechanism.

[0039] The perception module is composed of a binocular camera and an event camera, which uses computer vision technology to solve the position coordinates of the robot on the court, the space-time pose of the human body skeleton of the trainer, and the falling point information of the tennis ball and the position information of the scattered tennis ball, to provide effective data support for the analysis board of the electric control part. The analysis board is built-in with tennis action analysis and visual action feedback prompt algorithm, which drives the light strip and display screen in the feedback module to work according to the analysis result, to realize the real-time guidance to the trainer. Thus, the tennis training robot realizes embodied intelligence from perception to action.

[0040] Further, in the process of switching from the serving mode to the ball receiving mode, the whole friction wheel mechanism 2-1 moves and extends out of the robot body under the driving of the electric push rod 2-4-8; the pitch adjusting motor 2-1-4 is used to adjust the appropriate angle; the right swing arm 2-1-5 and the left swing arm 2-1-6 rotate counterclockwise under the driving of the right swing arm rudder 2-4-1 and the left swing arm rudder 2-4-2, so that the right friction wheel 2-1-1 and the left friction wheel 2-1-2 in the friction wheel mechanism 2-1 can approach and contact the tennis ball 6 on the ground, so as to throw the tennis ball 6 upward and make it fall into the ball basin 2-3-1 in the ball storage mechanism 2-3.

[0041] Further, in the ball receiving mode, the wheel-foot module controls the robot to perform the squatting action, and the ball receiving and sending module switches to the ball receiving mode and extends the friction wheel mechanism; the binocular camera in the perception module can accurately identify the position of the tennis ball in the court and send its coordinates to the electric control module; then, the electric control module directs the robot to move towards the tennis ball and makes the center of the friction wheel mechanism align with the tennis ball; under the action of friction, the tennis ball is thrown upward and falls smoothly into the ball basin of the ball storage mechanism, thereby completing the ball receiving process.

[0042] Further, the tennis training robot can perceive and feedback the kinematic performance of the tennis trainer's action; after the data from the binocular camera is processed by the existing human skeleton recognition algorithm in the analysis board, the position and angular velocity data of the trainer's hip, chest, shoulder, elbow and hand are obtained; let i represent the body part, i=1 represents the hip, i=2 represents the chest, i=3 represents the shoulder, i=4 represents the elbow, and i=5 represents the hand; ω i (t) represents the angular velocity vector of part i at time t, p i (t) represents the spatial position of part i at time t, v5(t) represents the spatial velocity of the hand at time t, the contribution velocity of each part to the hand is calculated and accumulated to the observed part i according to the following formula;

[0043]

[0044] wherein n is the cumulative variable, n=(1,...,i), and the final cumulative contribution velocity v Ai (t) is a scalar;

[0045] The tennis hitting action kinematic chain analysis mainly focuses on the force degree and force timing; the force degree CV i is calculated by the following formula:

[0046]

[0047] Let the time when the cumulative contribution velocity accumulated to part i reaches the peak value be t i , and the calculation method is

[0048]

[0049] The calculation method of the force timing is that the chest lags behind the hip by a time Δt1=t2-t1, the shoulder lags behind the chest by a time Δt2=t3-t2, and the elbow lags behind the shoulder by a time Δt3=t4-t3.

[0050] Further, the tennis training robot can visualize and feedback the kinematic chain of the trainer as follows: the leg lamp strip 4-2 feeds back the hip information, the chest lamp strip 4-3 feeds back the chest information, the large arm lamp strip 4-4 feeds back the shoulder information, and the small arm lamp strip 4-5 feeds back the elbow information; the color of the lamp strip represents the force degree, which is green for good RGB=(0, 255, 0), orange for better RGB=(255, 255, 0), yellow for poor RGB=(255, 50, 0), and red for serious poor RGB=(255, 0, 0), and the calculation method is as follows:

[0051]

[0052] wherein CV iRefReference value R representing the degree of force exertion iC , R iC , R iC CV1 is a fault tolerance value set artificially; if CV1 is higher than the reference value CV 1Ref , RGB1 is set as (0, 255, 0); if CV4 is lower than the reference value CV 4Ref , RGB4 is set as (0, 255, 0);

[0053] The coordination Δt of the force exertion timing of each part is fed back by the flowing mode of the light belt of each part i ; if there is a motion delay, the first light belt in the light belt of the part is lit up for f i seconds, the second light belt is lit up for 2f i seconds, and the third light belt is lit up for 3f i seconds; the flowing time interval f i of the light belt of the ith part of the whole light belt system is calculated as

[0054]

[0055] In the formula, Δt iRef represents the lag time of the part i, Δt iA , Δt iB , Δt iC represents the preset tolerance value; when f i = 0, it indicates that the LED of the area is always on; if Δt i is lower than the reference value Δt iRef , the flowing direction of the light belt is from the distal end to the proximal end of the body, prompting the trainer to slow down the timing of the part; otherwise, if Δt i is higher than the reference value Δt iRef , the flowing direction of the light belt is from the proximal end to the distal end of the body, prompting the trainer to speed up the timing of the part.

[0056] Compared with the prior art, the present application has the following beneficial effects:

[0057] 1. The tennis training robot of the present application combines the tasks that originally need to be completed by a ball launching robot and a ball picking robot into one, and realizes smooth switching of the receiving and launching functions through a cleverly designed mechanism. The robot can accurately control the trajectory of the tennis ball launching, and automatically find and recover the scattered tennis balls, thereby ensuring the full-automatic operation of the tennis training task, saving the labor cost of tennis training, and improving the stability and accuracy of the training.

[0058] 2.The tennis training robot has a flexible mechanical structure, which can accurately simulate the ball from different directions, speeds, ball heights and pitch angles. With its rapid lifting function, the robot can quickly lower its height after serving the ball to the trainer, preventing damage from being hit, effectively solving the problem of full-range feeding ball that traditional artificial coaches cannot achieve.

[0059] 3.The working principle of the body-intelligent tennis training robot gives the robot the ability to perceive, analyze and feedback, enabling it to interact with humans scientifically and efficiently during training, meeting the needs of tennis teaching and training in the new era. This working mode overcomes the limitations of traditional teaching and coach guidance, makes up for the shortcomings of existing motion analysis methods and training robot intelligence, and significantly improves training efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0060] In order to better describe and explain the technical solutions of the invention, clearly show the content and structure of the invention, the invention will be further described below in combination with the drawings:

[0061] Figure 1 The outer shape of the tennis training robot serving mode is shown in the axonometric view.

[0062] Figure 2 The outer shape of the tennis training robot receiving mode is shown in the axonometric view.

[0063] Figure 3 The wheel-foot module of the tennis training robot is shown in the axonometric view.

[0064] Figure 4 The receiving and serving module of the tennis training robot in the serving mode is shown in the axonometric view.

[0065] Figure 5 The receiving and serving module of the tennis training robot in the serving mode is shown in the top view.

[0066] Figure 6 The receiving and serving module of the tennis training robot in the receiving mode is shown in the axonometric view.

[0067] Figure 7 The launching mechanism of the receiving and serving module of the tennis training robot is shown in the axonometric view.

[0068] Figure 8 The ball storage mechanism of the receiving and serving module of the tennis training robot is shown in the axonometric view.

[0069] Figure 9 The electric control module of the tennis training robot is shown in the axonometric view.

[0070] Figure 10The tennis training robot module relationship diagram of the present application;

[0071] Figure 11 The geometric posture analysis principle diagram of the present application;

[0072] Label explanation:

[0073] Wheel-foot module 1; receiving and sending ball module 2; perception module 3; feedback module 4; electric control module 5; tennis ball 6;

[0074] Motor wheel 1-1; shank 1-2; foreleg 1-3; fore shank 1-4; thigh 1-5; foreleg motor 1-6; thigh motor 1-7; shell 1-8; strut 1-9; bearing 1-10; bottom plate 1-11;

[0075] Friction wheel mechanism 2-1; ball shooting mechanism 2-2; ball storage mechanism 2-3; receiving and sending ball switching mechanism 2-4;

[0076] Right friction wheel 2-1-1; left friction wheel 2-1-2; friction wheel support 2-1-3; pitch adjusting motor 2-1-4; right swing arm 2-1-5; left swing arm 2-1-6;

[0077] Ball shooting cylinder 2-2-1; ball pushing electric push rod 2-2-2; ball pushing block 2-2-3; push rod fixed block 2-2-4; push rod fixed frame 2-2-5;

[0078] Ball basin 2-3-1; stirring arm 2-3-2; stirring rudder 2-3-3;

[0079] Right swing arm rudder 2-4-1; left swing arm rudder 2-4-2; right guide rail 2-4-3; left guide rail 2-4-4; right rudder support 2-4-5; left rudder support 2-4-6; connecting block 2-4-7; electric push rod 2-4-8;

[0080] Binocular camera 3-1; event camera 3-2;

[0081] Display screen 4-1; leg lamp strip 4-2; chest lamp strip 4-3; large arm lamp strip 4-4; small arm lamp strip 4-5; arm rudder 4-6; arm 4-7;

[0082] Battery 5-1; control board 5-2; analysis board 5-3. DETAILED DESCRIPTION

[0083] In order to make the purpose, technical scheme and advantages of the embodiments of the present application more clear and more clear, the present application will be described in detail below with reference to the drawings.

[0084] In order to make the drawings simple and easy to understand, the reference numerals in the patent drawings adopt a two-level numbering rule, and not all parts of the robot are marked, but only the related parts of the important contents such as the structure and working principle described in the text are marked in detail.

[0085] In the invention described herein, unless there are specific provisions and limitations, the terms "mounting", "setting", "connection" and the like should be understood broadly. For example, they can include fixed connection, detachable connection, integral connection; can be mechanical connection, electrical connection; can be direct connection between two elements, or indirect connection through intermediate media, or even internal connection of elements. For those skilled in the art, the specific meaning of the above terms in the invention can be understood according to the specific context.

[0086] The present application provides an embodiment, a body-possessed intelligent tennis training robot, as shown in Figure 1 The tennis training robot of the present application shows its structural features in the perspective view of the serving mode. The robot adopts modular design, and the wheel-foot module 1 is located at the bottom of the robot, responsible for supporting, moving, steering and height adjustment. The serving-receiving module 2 is arranged in the middle of the robot, which can control the serving pitch angle and the ball speed of the robot in the serving mode; when switched to the receiving mode, it can collect the tennis on the ground into the ball basin 2-3-1 at the top of the robot. In addition, the robot is equipped with a perception module 3 composed of a binocular camera facing forward and an event camera facing backward, which aims to capture the positions of the human, the robot and the tennis team in space and time. The feedback module 4 includes a display screen 4-1 located at the top of the front of the robot and lamp strips distributed on the whole body of the robot, which are leg lamp strips 4-2, chest lamp strips 4-3, large arm lamp strips 4-4 and small arm lamp strips 4-5 respectively. At the same time, the arm servo 4-6 is used to control the stretching and tightening of the arm 4-7. The electric control module 5 is arranged inside the robot to ensure stable and efficient operation of the system.

[0087] As shown in Figure 2 In the receiving mode of the tennis training robot of the present application, the wheel-foot module 1 controls the robot to perform the squatting action, and the serving-receiving module 2 switches to the receiving mode and extends the friction wheel mechanism 2-1. The binocular camera 3-1 in the perception module 3 can accurately identify the position of the tennis 6 in the field and send its coordinates to the electric control module 5. Then, the electric control module commands the robot to move towards the tennis, and makes the center of the friction wheel mechanism 2-1 align with the tennis 6. Under the action of friction, the tennis is thrown upwards and successfully falls into the ball basin 2-3-1 of the ball storage mechanism 2-3, thus completing the receiving process.

[0088] As shown in Figure 3As shown in the isometric drawing, the structure of the wheel-foot module of the tennis training robot of the present invention is illustrated. The wheel-foot module consists of a pair of left and right wheels. A motor wheel 1-1 is mounted at the bottom, which rotates when powered on. The left and right wheels achieve omnidirectional movement by adjusting their speed difference. The motor wheel 1-1 is bolted to the lower leg 1-2. The middle of the lower leg 1-2 is connected to the bottom of the front leg 1-3 using bearings and pins to ensure relative rotation. The bottom of the front lower leg 1-4 is also connected to the front leg 1-3 using bearings and pins, while the top of the front lower leg 1-4 is bolted to the output shaft of the front leg motor 1-6. Furthermore, the upper part of the lower leg 1-2 is also connected to the lower part of the thigh 1-5 using bearings and pins, and the upper part of the thigh 1-5 is bolted to the output shaft of the thigh motor 1-7. The fixed ends of the front leg motor 1-6 and the thigh motor 1-7 are both secured to the outer shell 1-8 with bolts. The outer shell 1-8 is then connected to the support column 1-9 with bolts, and the support column 1-9 is mounted on the base plate 1-11. The motors of this pair of front legs and thighs can utilize the mature wheel-leg mechanism control principle to achieve height adjustment, thereby changing the overall height of the robot.

[0089] like Figures 4 to 8 As shown, the tennis training robot ball receiving and serving module 2 of the present invention mainly includes a friction wheel mechanism 2-1, a ball shooting mechanism 2-2, a ball storage mechanism 2-3, and a ball receiving and serving switching mechanism 2-4.

[0090] like Figure 4 This is an isometric view of the serve-receive module of the tennis training robot described in this invention in serve mode. The friction wheel mechanism 2-1 is at the front, and the ball-shooting mechanism 2-2 is located in the middle. The front of the ball-shooting mechanism 2-2 is close to the right friction wheel 2-1-1 and the left friction wheel 2-1-2 in the friction wheel mechanism. The serve-receive switching mechanism 2-4 is located at the rear and bottom, and is responsible for pushing the friction wheel mechanism 2-1 outward.

[0091] Figure 5 The working principle of friction wheel mechanism 2-1 is demonstrated. In friction wheel mechanism 2-1, both the right friction wheel 2-1-1 and the left friction wheel 2-1-2 are wheel motors, which can achieve high-speed rotation after being energized. Their motor ends are fixed to friction wheel bracket 2-1-3 by bolts. When the tennis ball 6 contacts the friction wheel, it will be launched rapidly. To adjust the launching angle of the tennis ball 6, one end of friction wheel bracket 2-1-3 is connected to the output shaft of pitch adjustment motor 2-1-4 by bolts, while the other end of the bracket is connected to right swing arm 2-1-5 by bearings. At the same time, pitch adjustment motor 2-1-4 is also fixed to left swing arm 2-1-6 by bolts.

[0092] Figure 5The working principle of the receiving and sending switching mechanism 2-4 is described in detail. The right swing arm steering gear 2-4-1 in the mechanism is connected to the right swing arm 2-1-5 in the friction wheel mechanism 2-1 through a bolt, and the left swing arm steering gear 2-4-2 is connected to the left swing arm 2-1-6 in the friction wheel mechanism 2-1 through a bolt. Under the drive of the steering gear, the friction wheel mechanism 2-1 can rotate around the steering gear shaft. The right swing arm steering gear 2-4-1 is fixed with the right steering gear support 2-4-5 through a bolt, and the right steering gear support 2-4-5 is fixed on the sliding block of the right guide rail 2-4-3 through a bolt to realize left and right movement; similarly, the left swing arm steering gear 2-4-2 is fixed with the left steering gear support 2-4-6 through a bolt, and the left steering gear support 2-4-6 is fixed on the sliding block of the left guide rail 2-4-4 through a bolt. The two ends of the connecting block 2-4-7 are connected with the right steering gear support 2-4-5 and the left steering gear support 2-4-6 through bolts, and the middle part of the connecting block 2-4-7 is fixed with the output end of the electric push rod 2-4-8 through a bolt. The rear end of the electric push rod 2-4-8 is fixed on the chassis 1-11 through a bolt.

[0093] As shown in Figure 6 , the isometric view of the tennis training robot receiving and sending ball module in the receiving ball mode. In the process of switching from the serving mode (see Figure 4 ) to the receiving mode (see Figure 6 ), the whole friction wheel mechanism 2-1 moves and extends out of the robot body under the drive of the electric push rod 2-4-8; the pitch adjusting motor 2-1-4 is used to adjust the appropriate angle; the right swing arm 2-1-5 and the left swing arm 2-1-6 rotate counterclockwise under the drive of the right swing arm steering gear 2-4-1 and the left swing arm steering gear 2-4-2, so that the right friction wheel 2-1-1 and the left friction wheel 2-1-2 in the friction wheel mechanism 2-1 can approach and contact the tennis ball 6 on the ground, so as to throw the tennis ball 6 upward and make it fall into the ball basin 2-3-1 in the ball storage mechanism 2-3.

[0094] Figure 7 The isometric view of the launching mechanism 2-2 in the tennis training robot receiving and sending ball module described by the present application. The top of the ball launching cylinder 2-2-1 in the ball launching mechanism 2-3 and the bottom of the ball basin 2-3-1 in the ball storage mechanism 2-3 are connected through corresponding bolts. The front end of the ball pushing electric push rod 2-2-2 is connected to the ball pushing block 2-2-3 through a bolt. After the tennis ball 6 in the storage mechanism 2-3 falls into the ball launching cylinder 2-2-1, it moves forward under the push of the ball pushing electric push rod 2-2-2, contacts the friction wheel and is launched. The outside of the ball pushing electric push rod 2-2-2 is in interference fit with the hole in the push rod fixing block 2-2-4, and the push rod fixing block 2-2-4 is connected to the ball launching cylinder 2-2-1 through a bolt.

[0095] Figure 8Axonometric view of the ball storage mechanism in the ball receiving and launching module of the tennis training robot. The ball basin 2-3-1 is installed on the top of the entire robot, where a large number of tennis balls 6 can be stored, and the bottom of the ball basin 2-3-1 is provided with a hole for the falling of only one tennis ball. In order to avoid the blockage of the falling ball hole by the accumulation of a large number of tennis balls, a stirring arm 2-3-2 is installed at the top of the hole, and under the action of the stirring rudder 2-3-3, the stirring arm 2-3-2 drives the tennis balls in the ball basin 2-3-1 to move at a low speed, so that they can freely fall into the launching barrel 2-2-1 when there is a vacant position in the barrel when the previous tennis ball is launched from the launching barrel 2-2-1, to fill the vacant position.

[0096] Figure 9 Axonometric view of the electric control module of the tennis training robot. The electric control module 5 is installed on the upper surface of the bottom plate 1-11, located at the rear of the robot. Two batteries 5-1 are arranged above the bottom plate 1-11, wherein the top of the right battery is equipped with a control panel 5-2, and the top of the left battery is equipped with an analysis panel 5-3.

[0097] The present application provides another embodiment, a working method of a body-intelligent tennis training robot, as shown in Figure 10As shown, the hardware composition and functional relationship diagram of the tennis training robot of the present application clearly shows the cooperative work of each module. The central electric control module in the figure is the control core of the whole system, responsible for motion control and perception feedback data processing. Among them, the main control board coordinates the work of the receiving and sending ball module and the wheel-foot module; the analysis board receives and processes the data from the perception module, and generates scientific decisions, and interacts with the trainer through the feedback module. The wheel-foot module realizes the height adjustment, movement and direction control of the whole robot. Four leg motors cooperatively control the bending degree of large and small legs to adjust the ball position and ensure the squatting posture when receiving the ball; the speed difference of left and right motor wheels determines the turning angle of the robot. When the speed of left and right motor wheels is the same, the robot moves straight; when there is a speed difference, the robot turns. The receiving and sending ball module has two working states of sending ball mode and receiving ball mode. In the sending ball mode, the tennis ball in the ball storage mechanism is released into the ball shooting cylinder, the electric push rod pushes the tennis ball to make it contact with the friction wheel, and the friction wheel generates a preset speed and rotation under the drive of the AC brushless motor. The rudder controls the pitch angle, the wheel-foot module controls the ball sending height and direction, and finally the tennis ball is thrown to the trainer according to the predetermined parameters. In the receiving ball mode, the receiving and sending switching mechanism is started, the electric push rod and the rudder work together to make the friction wheel mechanism move to the front and lower part of the robot, control the receiving speed, the pitch adjusting mechanism adjusts the receiving angle, the wheel-foot module controls the robot to squat and adjusts the direction, and finally the scattered tennis balls on the court are collected into the ball storage mechanism. The perception module is composed of a binocular camera and an event camera, which uses computer vision technology to solve the position coordinates of the robot on the court, the space-time pose of the trainer's human body skeleton, and the falling point information of the hit tennis ball and the position information of the scattered tennis ball, etc., to provide effective data support for the analysis board of the electric control part. The analysis board is built-in tennis action analysis and visual action feedback prompt algorithm, which drives the light strip and display screen in the feedback module to work according to the analysis result, realizing the real-time guidance to the trainer. Thus, the tennis training robot realizes the embodied intelligence from perception to action.

[0098] As Figure 11 shown is the perception and feedback method of the robot to the action posture of the tennis trainer. After the data from the binocular camera 3-1 is processed in the analysis board 5-3 using the existing human skeleton recognition algorithm, the action posture of the trainer is obtained. The whole tennis hitting action is divided into four links: preparation, lead, swing, and swing. In the preparation link, two features are used to describe the rationality of the action: the hip height ratio δ h (the ratio of hip height in preparation to standing), and the knee distance ratio δ L (the ratio of knee distance in preparation to standing). In the lead link, the angle α1 of the chest relative to the ball direction and the distance d1 of the hand relative to the chest position in the ball direction are used to describe whether the lead action is sufficient. In the swing phase, the angle α us between the chest and the upper arm at the hitting moment, and the angle αua The posture at the moment of hitting the ball is described. During the swing phase, only the residual speed needs to be released naturally, so no action posture analysis is performed. In the display screen 4-1 in the feedback module 4 of the robot, the characteristic values of each swing action of the trainer will be presented, and the trainer will be intuitively fed back in combination with the visual image mode shown in the analysis board 5-3. Figure 11

[0099] The robot can perceive and feedback the kinematic performance of the tennis trainer's action. After the data from the binocular camera 3-1 is processed by the existing human skeleton recognition algorithm in the analysis board 5-3, the position and angular velocity data of the trainer's hip, chest, shoulder (holding the racket hand), elbow (holding the racket hand), and hand are obtained. Let i represent the body part, i = 1 represent the hip, i = 2 represent the chest, i = 3 represent the shoulder, i = 4 represent the elbow, and i = 5 represent the hand. i (t) represents the angular velocity vector of part i at time t, p i (t) represents the spatial position of part i at time t, and v5(t) represents the spatial velocity of the hand at time t. According to the following formula, the contribution velocity of each part to the hand is calculated and accumulated to the observed part i

[0100]

[0101] where n is the cumulative variable, n = (1,...,i), and the final cumulative contribution velocity v Ai (t) is a scalar.

[0102] The tennis hitting action kinematic chain analysis mainly focuses on the force degree and force timing. The force degree CV i is calculated using the following formula:

[0103]

[0104] Let t i be the time when the cumulative contribution velocity accumulated to part i reaches the peak, and its calculation method is

[0105]

[0106] The calculation method of the force timing is that the chest lags behind the hip by time Δt1 = t2-t1, the shoulder lags behind the chest by time Δt2 = t3-t2, and the elbow lags behind the shoulder by time Δt3 = t4-t3.

[0107] As Figure 1 ​As shown, to visualize and provide feedback on the trainee's kinetic chain, light strip 4-2 on the legs provides feedback on hip information, light strip 4-3 on the chest provides feedback on chest information, light strip 4-4 on the upper arm provides feedback on shoulder information, and light strip 4-5 on the forearm provides feedback on elbow information. The color of the light strips indicates the degree of exertion: green for good (RGB=0,255,0), orange for relatively good (RGB=255,255,0), yellow for poor (RGB=255,50,0), and red for severely poor (RGB=255,0,0). The calculation method is as follows:

[0108]

[0109] In the formula, CV iRef R is a reference value indicating the degree of force exerted. iA R iB R iC This is a manually set tolerance value. Based on tennis experience, a higher hip CV (volume shift) improves stability and speed, while a lower elbow CV improves shot stability and prevents injury. Therefore, if CV1 is higher than the reference value CV... 1Ref If CV4 is higher than the reference value CV, then RGB1 is set to (0, 255, 0); 4Ref If the value is low, then RGB4 is set to (0,255,0).

[0110] The coordination Δt of the timing of force application in each part is fed back by the flow of light strips in each part. i If there is a delay in the action, the first light in that area of ​​the LED strip will light up. i Seconds later, the second light strip lit up, 2f i The third light strip lights up after 1 second. Therefore, the time interval f between the light strip movements at the i-th part of the entire light strip system is... i It can be calculated as

[0111]

[0112] In the formula, Δt i Δt represents the lag time of part i. iRef T represents the standard reference value. iA T iB T iC This represents the preset tolerance value. When f i =0 indicates that the LED in that area is constantly lit. If Δt i Below the reference value Δt iRef If the direction of the light strip flow is from the distal to the proximal end of the body (hip to foot, chest to hip, elbow to shoulder, hand to elbow), it indicates to the trainee that they need to slow down the timing of that part. Conversely, if Δt i Higher than the reference value Δt iRefIf the flow direction of the light belt is from the proximal end to the distal end of the body (from the foot to the hip, from the hip to the chest, from the shoulder to the elbow, and from the elbow to the hand), it indicates that the trainer needs to speed up the part of the body.

[0113] The above merely illustrates the specific embodiments of the present application, but the protection scope of the present application is not limited thereto, any modification, equivalent replacement, and improvement within the technical range disclosed by the present application and within the spirit and principle of the present application should be covered within the protection scope of the present application. Meanwhile, the contents not described in detail in the present specification are all the prior art known by the skilled in the art.

Claims

1. A working method of a body-integrated intelligent tennis training robot, the body-integrated intelligent tennis training robot comprising a wheel-foot module, a perception module, a ball-receiving and -launching module, a feedback module and an electronic control module; the method comprising the following steps: the electronic control module is responsible for motion control and perception feedback data processing; wherein, a main control board coordinates the work of the ball-receiving and -launching module and the wheel-foot module; an analysis board receives and processes data from the perception module, and generates scientific decisions, and carries out human-computer interaction with the trainer through the feedback module; the wheel-foot module realizes the height adjustment, movement and direction control of the whole robot; four leg motors cooperatively control the bending degree of large and small legs to adjust the ball launching position and ensure the squatting posture when receiving the ball; the speed difference of left and right motor wheels determines the turning angle of the robot; when the speed of the left and right motor wheels is the same, the robot moves straight; when there is a speed difference, the robot turns; the ball-receiving and -launching module has two working states of ball launching mode and ball receiving mode; in the ball launching mode, the tennis ball in the ball storage mechanism is released into the ball launching cylinder, the electric push rod pushes the tennis ball to make it contact with the friction wheel, the friction wheel generates a preset speed and rotation under the drive of the alternating current brushless motor, the rudder controls the pitch angle, the wheel-foot module controls the ball launching height and direction, and finally the tennis ball is launched to the trainer according to the predetermined parameters; in the ball receiving mode, the ball-receiving and -launching switching mechanism is started, the electric push rod and the rudder work together to make the friction wheel mechanism move to the front and lower part of the robot, control the ball receiving speed, the pitch adjusting mechanism adjusts the ball receiving angle, the wheel-foot module controls the robot to squat and adjusts the direction, and finally the tennis balls scattered on the court are collected into the ball storage mechanism; the perception module is composed of a binocular camera and an event camera, and uses computer vision technology to solve the position coordinates of the robot on the court, the space-time pose of the trainer's human body skeleton, and the falling point information of the hit tennis ball and the position information of the scattered tennis ball, to provide effective data support for the analysis board of the electronic control part; the analysis board is built-in with tennis action analysis and visual action feedback prompt algorithm, and drives the light strip and display screen in the feedback module to work according to the analysis result, to realize the real-time guidance to the trainer; thus, the tennis training robot realizes the body-integrated intelligence from perception to action. 2.The working method of the body-integrated intelligent tennis training robot according to claim 1, wherein: in the process of switching from the ball launching mode to the ball receiving mode, the whole friction wheel mechanism moves and extends out of the robot body under the drive of the electric push rod; the pitch adjusting motor is used to adjust the appropriate angle; the right swing arm and the left swing arm are driven by the right swing arm rudder and the left swing arm rudder to rotate counterclockwise, so that the right friction wheel and the left friction wheel in the friction wheel mechanism can approach and contact the tennis ball on the ground, so as to throw the tennis ball upward and make it fall into the ball basin in the ball storage mechanism. 3.The working method of the body-integrated intelligent tennis training robot according to claim 2, wherein: in the ball receiving mode, the wheel-foot module controls the robot to perform the squatting action, and at the same time the ball-receiving and -launching module switches to the ball receiving mode and extends the friction wheel mechanism; the binocular camera in the perception module can accurately identify the position of the tennis ball in the court and send its coordinates to the electronic control module. ​ ​ ​ ​ ​ ​ Subsequently, the electric control module instructs the robot to move towards the tennis and align the center of the friction wheel mechanism with the tennis; under the action of friction, the tennis is thrown upwards and falls into the ball basin of the ball storage mechanism, thus completing the ball collection process.

4. The method according to claim 1, characterized in that: The tennis training robot can sense and feedback the kinematic performance of the tennis trainer's action. After analyzing the data from the binocular camera in the analysis board using the existing human skeleton recognition algorithm, the position and angular velocity data of the trainer's hip, chest, shoulder, elbow and hand are obtained. Let i represent the body part, i=1 represents the hip, i=2 represents the chest, i=3 represents the shoulder, i=4 represents the elbow, and i=5 represents the hand. i (t) represents the angular velocity vector of part i at time t, p i (t) represents the spatial position of part i at time t, v5(t) represents the spatial velocity of the hand at time t, and the contribution velocity of each part to the hand is calculated and accumulated to the observation part i according to the following formula; where n is a cumulative variable, n = (1,..., i), and the final cumulative contribution velocity v Ai (t) is a scalar; The tennis stroke motion chain analysis mainly focuses on the force degree and force timing; the force degree CV i The following formula is used for calculation: Let t be the time at which the cumulative contribution velocity up to site i reaches its peak value i The calculation method is The calculation method of the timing of the force is that the chest lags behind the hips by a time Δt1=t2-t1, the shoulder lags behind the chest by a time Δt2=t3-t2, and the elbow lags behind the shoulder by a time Δt3=t4-t3.

5. The method according to claim 4, characterized in that: The visual feedback of the tennis training robot to the movement chain of the trainer is that the leg light band is used to feedback the hip information, the chest light band is used to feedback the chest information, the large arm light band is used to feedback the shoulder information, and the small arm light band is used to feedback the elbow information; the color of the light band is used to represent the force degree, which is green for good RGB=(0, 255, 0), orange for better RGB=(255, 255, 0), yellow for poor RGB=(255, 50, 0), and red for serious poor RGB=(255, 0, 0); the calculation method is as follows: wherein CV iRef represents a reference value of the degree of force, R iA , R iB , R iC is a fault tolerance value set artificially; if CV1 is higher than the reference value CV 1Ref , RGB1 is set as (0, 255, 0); if CV4 is lower than the reference value CV 4Ref , RGB4 is set as (0, 255, 0). The coordination of the force exertion timing of each part is fed back by the flowing of the light belt of each part i If there is action delay, the first light belt in the part light belt lights up f i seconds, the second light belt lights up 2f i seconds, and the third light belt lights up; the light belt flowing time interval f i of the i-th part of the whole light belt system is calculated as where Δt i represents the lag time of site i, Δt iRef represents the standard reference value, T iA , T iB , T iC represents the preset tolerance value; when f i = 0 indicates that the LED of the region is always on; if Δt i is lower than the reference value Δt iRef , the flow direction of the light belt is from the distal end to the proximal end of the body, prompting the trainer to slow down the timing of the part; otherwise, if Δt i is higher than the reference value Δt iRef , the flow direction of the light belt is from the proximal end to the distal end of the body, prompting the trainer to speed up the timing of the part.

6. The method according to claim 1, characterized in that: The ball collection and transmission module is located above the wheel-foot module, can control the pitching angle and the ball speed of the robot in the ball serving mode, and can collect the tennis in the ball collection mode; The feedback module is arranged on the front of the robot and visually feedbacks the hitting action of the trainer through the light band and the display screen; The electric control module includes a main control board, an analysis board and a power supply assembly, is responsible for the motion control and the sensing feedback data processing; the main control board coordinates the work of the ball collection and transmission module and the wheel-foot module; the analysis board receives and processes the data from the sensing module, generates scientific decisions, and performs human-computer interaction with the trainer through the feedback module.

7. The method according to claim 1, characterized in that: The ball collection and transmission module mainly includes a friction wheel mechanism, a ball shooting mechanism, a ball storage mechanism and a ball collection and transmission switching mechanism; The friction wheel mechanism drives the friction wheel to rotate at different speeds and directions through a pair of AC brushless motors, realizes the ball serving mode of outward shooting and the ball collection mode of inward recovery, controls the angle of the friction wheel fixed support through a pitching adjustment motor, realizes the shooting angle adjustment in the ball serving state and the control of the angle of the friction wheel in the ball collection and serving state switching; The ball shooting mechanism includes a three-way ball shooting cylinder, can push the tennis in the ball shooting cylinder to contact with the friction wheel and shoot out under the action of the ball pushing electric push rod; after the push rod is withdrawn, the next tennis to be shot falls down and enters the shooting preparation state under the action of gravity; The ball storage mechanism is located at the top of the robot and the collected tennis falls into the ball storage bin; The ball collection and transmission switching mechanism gives the friction wheel the ability to move forward and backward and swing, and the friction wheel mechanism is stored in the abdomen of the robot in the ball serving mode. In the ball collection mode, the friction wheel mechanism is driven by the push rod and the rudder to extend from the robot's abdomen and swing to the ground level to pick up the ground tennis ball.

8. The method of claim 1, wherein the robot is an embodied intelligent tennis training robot. The stirring arm is arranged in the ball storage bin, and the bottom tennis ball keeps a slow moving state under the action of the stirring arm in the ball storage bin to prevent the tennis ball from being blocked.

9. The method of claim 1, wherein the robot is an embodied intelligent tennis training robot. The wheel-foot module is located at the bottom of the robot and is responsible for supporting, moving, steering, and height adjustment. The perception module is composed of a double-eye camera facing forward and an event camera facing backward, which is used to capture the action posture of the trainer, the position of the robot in the court, and the landing point information of the tennis ball. The feedback module includes a display screen located at the top of the front of the robot and lamp strips distributed on the entire body of the robot, which are leg lamp strips, chest lamp strips, large arm lamp strips, and small arm lamp strips.

10. The method of claim 1, wherein the robot is an embodied intelligent tennis training robot. The right swing arm rudder of the receiving and transmitting switching mechanism is connected with the right swing arm in the friction wheel mechanism through a bolt, and the left swing arm rudder is connected with the left swing arm in the friction wheel mechanism through a bolt; under the drive of the rudder, the friction wheel mechanism can rotate around the rudder shaft; the right swing arm rudder is fixed with the right rudder support through a bolt, and the right rudder support is fixed on the slider of the right guide rail through a bolt to realize left and right movement; similarly, the left swing arm rudder is fixed with the left rudder support through a bolt, and the left rudder support is fixed on the slider of the left guide rail through a bolt; the two ends of the connecting block are connected with the right rudder support and the left rudder support through bolts, the middle part of the connecting block is fixed with the output end of the electric push rod through a bolt, and the rear end of the electric push rod is fixed on the bottom plate through a bolt.

Images