Intelligent tennis ball picking robot

Abstract

The invention discloses an intelligent tennis ball picking robot, and particularly relates to the technical field of robots, in the vision aspect, an OpenMV vision module is used, environment information and test information are collected, the running speed of the robot is predicted through a time sequence model, PID control is used for a chassis, the real running speed of the robot is made to be consistent with the predicted running speed, and the running speed of the robot is guaranteed. Specific parameter adjustment is carried out in the aspect of tennis ball picking, so that the tennis ball is smoothly operated in a tennis court scene; an HC-08 Bluetooth module is additionally arranged, wireless communication with other devices can be achieved, a user can have better experience, and the robot can be controlled by himself / herself to grab a tennis ball; and the predicted driving speed is determined, the maximum value and the minimum value of the angle of the steering engine correspond to the corresponding predicted driving speed, then the steering engine is controlled to rotate to grab the tennis ball, and the tennis ball grabbing is more accurate.

Description

Technical Field

[0001] The present invention relates to the field of robot technology, and more specifically, to an intelligent tennis ball picking robot. Background Art

[0002] In terms of machine vision, domestic universities and research institutions have made a series of progress through introduction and independent research and development. For example, the intelligent tennis ball picking robot developed by a university uses a binocular vision system to synthesize the position information of the tennis ball in three-dimensional space through the images captured by two cameras, which improves the accuracy of positioning and provides accurate guidance for the movement of the robot arm.

[0003] China has also made remarkable achievements in autonomous navigation technology. Robots developed by some companies use the data fusion of inertial measurement units (IMU) and encoders to achieve autonomous positioning technology that does not rely on external positioning systems. At the same time, combined with classic path planning algorithms, efficient and stable movement strategies are achieved.

[0004] In terms of human-computer interaction, domestic researchers are also actively exploring new interaction modes. They try to apply touch screen and voice recognition technology to intelligent tennis ball picking robots, allowing users to control the robot's behavior by touching the screen or using voice commands, improving the convenience and affinity of operation.

[0005] In terms of ball picking methods, domestic intelligent tennis ball picking robots mostly adopt a manually pushed cart design, as well as a traditional ball picking tube and manually controlled robotic arms. Although these designs are not as automated as foreign products, they have certain advantages in cost control and market penetration. With the continuous advancement of technology and the reduction of costs, the ball picking methods of domestic intelligent tennis ball picking robots will gradually develop in the direction of automation and intelligence.

[0006] The prior art has the following deficiencies:

[0007] The robot control algorithm is not precise enough, which increases its power consumption and reduces its working efficiency and working duration; at the same time, the robot speed control is relatively rough.

[0008] In order to solve the above-mentioned problem, a technical solution is now provided. Summary of the invention

[0009] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an intelligent tennis ball picking robot to solve the problems raised in the above-mentioned background technology.

[0010] To achieve the above object, the present invention provides the following technical solutions:

[0011] An intelligent tennis ball picking robot comprises a data acquisition module, an infrared obstacle avoidance module, an analysis module, a motor control module, a mechanical arm control module and a Bluetooth module;

[0012] The data acquisition module collects environmental information and test information, and transmits the environmental information and test information to the analysis module;

[0013] The infrared obstacle avoidance module collects obstacle information, controls the robot's steering, and transmits the obstacle information to the analysis module;

[0014] The analysis module receives the environmental information and test information transmitted by the data acquisition module and the infrared obstacle avoidance module, analyzes and determines the predicted driving speed, and transmits the predicted driving speed to the motor control module;

[0015] The motor control module confirms the predicted driving speed and controls the motor output through the PID algorithm to make the robot's driving speed consistent with the predicted driving speed;

[0016] The robot arm control module determines the predicted driving speed. The maximum and minimum values ​​of the servo angle correspond to the corresponding predicted driving speed, thereby controlling the rotation of the servo.

[0017] Bluetooth module, connects the robot to the client.

[0018] In a preferred embodiment, the environmental information acquired by the data acquisition module includes road condition information and tennis information, and the test information includes the driving speed of the robot.

[0019] In a preferred embodiment, the road condition information includes road surface dryness and smoothness, and the tennis ball information includes the distance between the tennis ball and the robot and the moving speed of the tennis ball.

[0020] In a preferred embodiment, the infrared obstacle avoidance module determines obstacle information, and the obstacle information includes the size and distance of the obstacle.

[0021] In a preferred embodiment, the obstacle information and road condition information are determined using a logistic regression algorithm: L = 1-e -(a*gz+b*pd+c*za) Calculate the road condition information index; where L represents the road condition information index, gz represents the road surface dryness, pd represents the road surface smoothness, za represents the distance between the obstacle and the robot, and a, b, and c are the regression coefficients of the variables;

[0022] Determine tennis information through the logistic regression formula: W = 1-e -(d*jl+f*sd) Calculate the tennis motion index; where W represents the tennis motion index, jl represents the distance between the tennis ball and the robot, sd represents the tennis motion speed, and d and e are the regression coefficients of the variables.

[0023] In a preferred embodiment, the test information, tennis sport index and road condition information index are determined to use a time series model to predict the robot's driving speed. The time series model uses an ARIMAX model. The specific steps are as follows:

[0024] Step S1, obtaining prediction data;

[0025] Step S2, establishing an ARIMAX model;

[0026] Step S3, using the maximum likelihood estimation (MLE) method to estimate the ARIMAX model parameters;

[0027] Step S4, verifying the effect of the fitting model and checking the goodness of fit of the model by residual analysis;

[0028] Step S5, using the fitted model to predict the future driving speed, and taking the maximum value of the prediction result as the predicted driving speed.

[0029] In a preferred embodiment, the exogenous variables in the ARIMAX model include a road condition information index and a tennis sport index, and the ARIMAX model formula is:

[0030] In the formula, X 1,t is the traffic information index at time point t, β1 is the coefficient of the traffic information index, X 2,t is the tennis movement index at time point t, β2 is the coefficient of the tennis movement index, is the first-order autoregressive parameter, θ1 is the first-order velocity average parameter, μ, , θ1, β1, and β2 are calculated and obtained by the maximum likelihood estimation method.

[0031] In a preferred embodiment, the motor control module determines the predicted driving speed and controls the motor output through a PID algorithm so that the robot driving speed is consistent with the predicted driving speed.

[0032] In a preferred embodiment, the robotic arm control module determines the predicted driving speed value of the robot, the maximum and minimum values ​​of each servo angle correspond to the corresponding predicted driving speed values, and the servo is rotated to grab the tennis balls. When a certain number of tennis balls or tennis balls are grabbed, the specific number is set according to the maximum limit that the robot frame can load; each time the grab is made, the count is increased by one. When the number of grabs is greater than the maximum value set by the system, the grabbing operation will no longer be performed, and information will be sent to the Bluetooth communication module to notify the user to clean the tennis balls in the frame.

[0033] In a preferred embodiment, the use process of an intelligent tennis ball picking robot is as follows:

[0034] Step A1: System initialization, camera turned on, information collection module works;

[0035] Step A2: Check whether there is a tennis ball in the image captured by the camera. If not, the robot rotates twice in place to search for the tennis ball.

[0036] Step A3: Determine the position of the tennis ball, and the robot moves toward the tennis ball;

[0037] Step A4: The analysis module determines the test data and the measured environment data to calculate the predicted driving speed of the robot;

[0038] Step A5: The motor control module determines the predicted driving speed of the robot and controls the motor output through the PID algorithm so that the driving speed of the robot is consistent with the predicted driving speed;

[0039] Step A6: If an obstacle is encountered during driving, the infrared obstacle avoidance module controls the robot to turn, avoid the obstacle, and search for the tennis ball again;

[0040] Step A7: When the robot moves near the tennis ball, the robotic arm control module controls the robotic arm to pick up the tennis ball, put it into the box, increase the number by one, and determine whether the box is full. If not, continue to pick up tennis balls. If full, notify the user.

[0041] Technical effects and advantages of the present invention:

[0042] The present invention uses an OpenMV vision module in the visual aspect to collect environmental information and test information, and predicts the robot's driving speed through a time series model. The chassis of the present invention uses PID control to make the robot's actual driving speed consistent with the predicted driving speed, and performs specific parameter adjustment in picking up tennis balls to enable it to operate smoothly in a tennis court scene; the present invention is accompanied by an HC-08 Bluetooth module, which can wirelessly communicate with other devices to give users a better experience, and can control the robot to grab tennis balls by itself; and by determining the predicted driving speed, the maximum and minimum values ​​of the servo angle correspond to the corresponding predicted driving speed, thereby controlling the servo to rotate and grab the tennis balls. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0044] Figure 1 This is a schematic diagram of the internal functional structure of an intelligent tennis ball picking robot of the present invention;

[0045] Figure 2 This is a PID algorithm flow chart of an intelligent tennis ball picking robot of the present invention;

[0046] Figure 3 This is a working diagram of a servo engine of an intelligent tennis ball picking robot according to the present invention;

[0047] Figure 4 The present invention is an operation flow chart of an intelligent tennis ball picking robot. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] The present invention uses an OpenMV vision module in the visual aspect to collect environmental information and test information, and predicts the robot's driving speed through a time series model. The chassis of the present invention uses PID control to make the robot's actual driving speed consistent with the predicted driving speed, and performs specific parameter adjustment in picking up tennis balls to enable it to operate smoothly in a tennis court scene. The present invention is accompanied by an HC-08 Bluetooth module, which can communicate wirelessly with other devices; and by determining the predicted driving speed, the maximum and minimum values ​​of the servo angle correspond to the corresponding predicted driving speed, thereby controlling the servo to rotate and grab the tennis ball.

[0050] Example 1

[0051] The present invention discloses an intelligent tennis ball picking robot. Figure 1 , Figure 2 As shown, it includes a data acquisition module, an infrared obstacle avoidance module, an analysis module, a motor control module, a robotic arm control module, and a Bluetooth module.

[0052] The data acquisition module collects environmental information and test information, and transmits the environmental information and test information to the analysis module;

[0053] The infrared obstacle avoidance module collects obstacle information, controls the robot's steering, and transmits the obstacle information to the analysis module;

[0054] The analysis module receives the environmental information and test information transmitted by the data acquisition module and the infrared obstacle avoidance module, analyzes and determines the predicted driving speed, and transmits the predicted driving speed to the motor control module;

[0055] The motor control module confirms the predicted driving speed and controls the motor output through the PID algorithm to make the robot's driving speed consistent with the predicted driving speed;

[0056] The robot arm control module determines the predicted driving speed. The maximum and minimum values ​​of the servo angle correspond to the corresponding predicted driving speed, thereby controlling the rotation of the servo.

[0057] Bluetooth module, connects the robot to the client.

[0058] The data acquisition module collects test information including the robot's moving speed; environmental information includes road condition information and tennis ball information; at the beginning, the robot rotates in place to capture whether the tennis ball appears in the picture, and if it does not appear in the picture, it rotates in place twice.

[0059] Road condition information includes the dryness and smoothness of the road surface, including laser or infrared sensors: these sensors scan the road surface through infrared or lasers, and infer the dryness of the road surface by the difference between the road surface temperature and the ambient temperature; when the road surface humidity is high, the surface temperature is low, and when it is dry, the surface temperature is high; based on this, the dryness of the road surface can be derived; tilt sensors usually measure the change in the direction of gravity through built-in accelerometers or gyroscopes, thereby inferring the angle between the object and the horizontal plane, and then deriving the smoothness of the road surface.

[0060] The test data includes the speed of the robot at each time point. The speed of the robot at each time point is obtained by installing speed sensors and clocks.

[0061] The tennis information includes the distance between the tennis ball and the robot, the speed of the tennis ball, and if the tennis ball is moving away from the robot, the speed of the tennis ball is positive; if the tennis ball is moving towards the robot, the speed of the tennis ball is negative. It can be obtained through the OpenMV sensor; the OpenMV sensor is an open source hardware vision sensor designed for embedded vision applications, with strong image processing capabilities and low power consumption. The core of OpenMV is a microcontroller-based camera module, which is commonly used in applications such as machine vision, robot control, object recognition, and color tracking. Through its powerful image processing capabilities and Python programming interface, users can achieve various visual tasks without the need for a high-performance PC or complex hardware.

[0062] The infrared obstacle avoidance module is used to avoid obstacles encountered by the robot during driving; it controls the infrared transmitter to emit infrared light (usually invisible infrared light) to the front or surrounding environment. The infrared receiver then receives the reflected infrared light and measures the intensity of the reflected light and the time it takes to return (such as time of flight, ToF: Time of Fl ight technology). By analyzing these data, it is possible to infer obstacle information, including the size and distance of the obstacle, and then control the robot to turn and avoid the obstacle; at the same time, the obstacle information is sent to the analysis and processing module.

[0063] The analysis and processing module predicts and analyzes the robot's driving speed through a time series model to determine the robot's maximum driving speed; the maximum speed at which the robot can travel represents the maximum speed at which the robot can move safely and is the most efficient driving speed. The maximum driving speed is used as the robot's predicted driving speed; and the robot's predicted driving speed is sent to the motor control module.

[0064] The time series model used in this embodiment is an ARIMAX model, which specifically includes the following steps:

[0065] Step S1, obtaining prediction data;

[0066] Step S2, establishing an ARIMAX model;

[0067] Step S3, using the maximum likelihood estimation (MLE) method to estimate the ARIMAX model parameters;

[0068] Step S4, verifying the effect of the fitting model and checking the goodness of fit of the model by residual analysis;

[0069] Step S5, using the fitted model to predict the future driving speed, and taking the maximum value of the prediction result as the predicted driving speed.

[0070] More specifically, the prediction data includes test data and measurement environment data, wherein the test data includes the robot's driving speed corresponding to each time point, and the measurement environment data includes a road condition information index and a tennis movement index.

[0071] The road condition information index is used to reflect the road condition information of the tennis court. The comprehensive road condition information and obstacle information can be calculated by the logistic regression formula: L = 1-e -(a*gz+b*pd+c*za) ; Wherein L represents the road condition information index, gz represents the dryness of the road surface, wherein the higher the road surface dryness, the higher the road condition information index, and vice versa; pd represents the smoothness of the road surface, wherein the higher the road surface smoothness, the higher the road condition information index, and vice versa; za represents the distance between the obstacle and the robot, wherein the greater the distance between the obstacle and the robot, the higher the road condition information index, and vice versa; a, b, and c are the regression coefficients of the variables, all greater than zero.

[0072] The tennis movement index is used to reflect the movement of tennis. It is calculated using a logistic regression algorithm combined with tennis information: W = 1-e -(d*jl+f*sd) ; Where W represents the tennis movement index, jl represents the distance between the tennis ball and the robot, where the farther the tennis ball is from the robot, the higher the tennis movement index, and vice versa; sd represents the tennis movement speed, where the higher the tennis movement speed, the higher the tennis movement index, and vice versa; d and e are the regression coefficients of the variables, both greater than zero.

[0073] To be more specific, the basic form of the ARIMAX model is:

[0074] In the formula, y t is the velocity at time point t; μ is a constant term, representing the intercept of the model; p is the order of the AR part, representing the autoregressive effect of the past p time points; Φ i is the parameter of the AR part, which indicates the relationship between the predicted value and its past value; q is the parameter of the MA part, which indicates the influence of the prediction error at the past q time points; θ j is the parameter of the MA part, which indicates the impact of the error term on the current value; r is the order of the exogenous variable part, which indicates the lag order of the exogenous variable; x t-k is the exogenous variable x t The value at time tk (can also be a combination of multiple exogenous variables); β k is the coefficient of the exogenous variable, indicating the impact of the exogenous variable on the target variable; ∈ t is the error term at time point t.

[0075] In this embodiment, the exogenous variables include the road condition information index L and the tennis sport index W, that is, the AR IMAX model of this embodiment is:

[0076] In the formula, X 1,t is the traffic information index at time point t, β1 is the coefficient of the traffic information index, X 2,t is the tennis movement index at time point t, β2 is the coefficient of the tennis movement index, is the first-order autoregressive parameter, and θ1 is the first-order velocity average parameter.

[0077] It should be noted that in step A3, μ, , θ1, β1, β2 are calculated by the maximum likelihood estimation (MLE) method. The specific steps are as follows:

[0078] Error term ∈ t :

[0079] Then the likelihood function can be written as:

[0080]

[0081] Taking the logarithm of the likelihood function gives the log-likelihood function:

[0082]

[0083] By maximizing the log-likelihood function, we can obtain the parameter estimates μ, , θ1, β1, β2.

[0084] The motor control module determines the predicted driving speed, which is the predicted driving speed for the target speed that the robot wants to achieve. However, the robot will be affected by friction and other factors during its movement, so the actual speed cannot reach the target speed. Therefore, the PID algorithm is needed to make the robot's driving speed reach the predicted driving speed.

[0085] PID algorithm, that is, proportional-integral-differential algorithm, such as Figure 2 As shown in Figure 1, it is a feedback control method widely used in industrial control systems. The PID controller adjusts the control output by calculating the deviation between the set point (expected value) and the actual measured value to make the system reach or maintain the desired state. The PID algorithm consists of three main parts: proportional (P), integral (I) and differential (D). Each part has its specific role and parameters, which are described in detail below.

[0086] Proportional (P) control:

[0087] Proportional control is the simplest form of control and adjusts the control output based on the difference between the set point and the actual measured value.

[0088] The proportional coefficient (Kp) determines the sensitivity of the control output to deviation. A larger Kp can speed up the system response, but may cause overshoot and oscillation; a smaller Kp may slow down the system response.

[0089] Integral (I) control:

[0090] Integral control takes into account the effect of deviation accumulation, that is, the historical information of the error.

[0091] Integration time constant (T i ) determines the strength of the integral action. The smaller T i It will enhance the integral action and help eliminate static errors, but it may cause the transient process of the system to become longer.

[0092] Derivative (D) control:

[0093] Derivative control focuses on the changing trend of the deviation, that is, the future information of the error.

[0094] Differential time constant (T d ) determines the strength of the differential action. Appropriate T d The future behavior of the system can be predicted and adjustments can be made in advance to reduce overshoot and oscillation.

[0095] The basic formula of the PID algorithm is as follows:

[0096]

[0097] Where u(t) is the control output; e(t) is the deviation, that is, the difference between the set point and the measured value; K d is the proportionality coefficient; K i is the integration coefficient, usually expressed as K i =K p / T i ; K d is the differential coefficient, usually expressed as K d =K d *T d .

[0098] First, initialize and set K p ,K i , K d , set the target speed or position setpoint, enter the loop. Read the current speed or position current_value,

[0099] Calculate the deviation: error = setpoint - current_value;

[0100] Calculate the proportional term: p = K p *error;

[0101] Update the integral term: I = I + K i *error;

[0102] Calculate the differential term: D = K d *(error-previous error);

[0103] Control output: control_signal = P + I + D;

[0104] Then limit the control output value to within the allowed range, apply the control signal to the motor (e.g. adjust the predicted driving speed duty cycle), wait for the sampling period, record the current error as the “previous error” for the next iteration, and finally determine whether to stop the loop.

[0105] In the robot control module, set the timer, set the pre-scaling factor to 167, set the main counter to 999, and the total clock (HCLK) set for this system is 168MHz. Therefore, the cycle is (167+1)*(999+1) / 168000000=1ms. Secondly, wait for the robot to follow the tennis ball to an appropriate distance. Once the grab variable is set to 1, it means entering the grab process. The servo will execute in turn according to the rotation angle, and the mechanical claw will grab. During the rotation process, it is necessary to add a loop delay to adjust the angle. Do not rotate a large angle at once to prevent excessive inertia from causing the robot arm to shake and thus shake as a whole. After the grab is completed, swing it back and put it into the box.

[0106] Determine the predicted driving speed of the robot. The predicted driving speed is the driving speed of the robot. The maximum and minimum values ​​of each servo angle correspond to the corresponding predicted driving speed. Secondly, add a delay of about 10ms to the rotation function, so that the servo does not rotate too fast and looks smooth. The servo rotation flow chart is as follows: Figure 3 shown.

[0107] When grabbing a certain number of tennis balls or tennis balls, the specific number is set according to the maximum limit that the robot frame can load; the count is increased by one each time the robot grabs. When the number of grabs is greater than the maximum value set by the system, the grabbing operation will no longer be performed, and information will be sent to the Bluetooth communication module to notify the user to clear the tennis balls in the frame.

[0108] The Bluetooth communication module transmits the tennis movement information collected by the data acquisition module through the HC-08 Bluetooth module, so that the user can see the position of the tennis ball through the user terminal, and can directly control the motor and the robot arm through the user terminal to realize the movement of the robot and the grabbing of the tennis ball; so that users have a good interactive experience. When the box is full, a prompt message will be sent to the client. At the same time, if the robot fails, a message will also be sent to the user terminal.

[0109] The use process of the above-mentioned intelligent tennis ball picking robot is as follows: Figure 4 As shown, the specific steps are:

[0110] Step A1: Initialize the system and turn on the camera;

[0111] Step A2: Check whether there is a tennis ball in the image captured by the camera. If not, the robot rotates twice in place to search for the tennis ball.

[0112] Step A3: Determine the position of the tennis ball, and the robot moves toward the tennis ball;

[0113] Step A4: Determine the test data and the measured environment data to calculate the predicted driving speed of the robot;

[0114] Step A5: Determine the predicted driving speed of the robot, and control the motor output through the PID algorithm so that the driving speed of the robot is consistent with the predicted driving speed;

[0115] Step A6: If an obstacle is encountered during driving, the robot is controlled to turn to avoid the obstacle and search for the tennis ball again;

[0116] Step A7: When moving near the tennis ball, control the robotic arm to pick up the tennis ball, put it into the box, increase the number by one, and determine whether the box is full. If not, continue to pick up tennis balls. If full, notify the user.

[0117] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0118] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0119] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0120] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0121] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An intelligent tennis ball picking robot, characterized in that: Including data acquisition module, infrared obstacle avoidance module, analysis module, motor control module, robotic arm control module, Bluetooth module; Data collection module, collecting environmental information and test information; and transmits environmental information and test information to the analysis module; The infrared obstacle avoidance module collects obstacle information, controls the robot's steering, and transmits the obstacle information to the analysis module; The analysis module receives the environmental information and test information transmitted by the data acquisition module and the infrared obstacle avoidance module, analyzes and determines the predicted driving speed, and transmits the predicted driving speed to the motor control module; The motor control module confirms the predicted driving speed and controls the motor output through the PID algorithm to make the robot's driving speed consistent with the predicted driving speed; The robot arm control module determines the predicted driving speed. The maximum and minimum values ​​of the servo angle correspond to the corresponding predicted driving speed, thereby controlling the rotation of the servo. Bluetooth module, connects the robot to the client.

2. The intelligent tennis ball picking robot according to claim 1, characterized in that: The environmental information acquired by the data acquisition module includes road condition information and tennis ball information, and the test information includes the robot's driving speed.

3. The intelligent tennis ball picking robot according to claim 2, characterized in that: The road condition information includes the dryness and smoothness of the road surface, and the tennis ball information includes the distance between the tennis ball and the robot and the speed of the tennis ball.

4. The intelligent tennis ball picking robot according to claim 1, characterized in that: The infrared obstacle avoidance module determines obstacle information, which includes the size and distance of the obstacle.

5. The intelligent tennis ball picking robot according to claim 1, characterized in that: Determine obstacle information and road condition information using logistic regression algorithm: L = 1-e -(a*gz+b*pd+c*za) Calculate the road condition information index; where L represents the road condition information index, gz represents the road surface dryness, pd represents the road surface smoothness, za represents the distance between the obstacle and the robot, and a, b, and c are the regression coefficients of the variables; Determine tennis information through the logistic regression formula: W = 1-e -(d*jl+f*sd) Calculate the tennis motion index; where W represents the tennis motion index, jl represents the distance between the tennis ball and the robot, sd represents the tennis motion speed, and d and e are the regression coefficients of the variables.

6. The intelligent tennis ball picking robot according to claim 1, characterized in that: The analysis module determines the test information, tennis movement index and road condition information index and uses the time series model to predict the robot's driving speed. The time series model uses the ARIMAX model. The specific steps are as follows: Step S1, obtaining prediction data; Step S2, establishing an ARIMAX model; Step S3, using the maximum likelihood estimation (MLE) method to estimate the ARIMAX model parameters; Step S4, verifying the effect of the fitting model and checking the goodness of fit of the model by residual analysis; Step S5, using the fitted model to predict the future driving speed, and taking the maximum value of the prediction result as the predicted driving speed.

7. The intelligent tennis ball picking robot according to claim 6, characterized in that: The exogenous variables in the ARIMAX model include the road condition information index and the tennis sport index. The ARIMAX model formula is: Where, X 1,t is the traffic information index at time point t, β1 is the coefficient of the traffic information index, X 2,t is the tennis movement index at time point t, β2 is the coefficient of the tennis movement index, is the first-order autoregressive parameter, θ1 is the first-order velocity average parameter, μ, θ1, β1, and β2 are calculated using the maximum likelihood estimation method.

8. The intelligent tennis ball picking robot according to claim 7, characterized in that: The motor control module determines the predicted driving speed and controls the motor output through the PID algorithm so that the robot's driving speed is consistent with the predicted driving speed.

9. The intelligent tennis ball picking robot according to claim 1, characterized in that: The robotic arm control module determines the predicted driving speed of the robot. The maximum and minimum values ​​of each servo angle correspond to the corresponding predicted driving speed. The servo is rotated to grab the tennis balls. When a certain number of tennis balls or tennis balls are grabbed, the specific number is set according to the maximum load limit of the robot frame. The count is increased by one each time the robot grabs the tennis balls. When the number of grabs is greater than the maximum value set by the system, the grabbing operation will no longer be performed, and information will be sent to the Bluetooth communication module to notify the user to clean up the tennis balls in the frame.

10. The intelligent tennis ball picking robot according to claim 1, characterized in that: The use process of an intelligent tennis ball picking robot is as follows: Step A1: System initialization, camera turned on, information collection module works; Step A2: Check whether there is a tennis ball in the image captured by the camera. If not, the robot rotates twice in place to search for the tennis ball. Step A3: Determine the position of the tennis ball, and the robot moves toward the tennis ball; Step A4: The analysis module determines the test data and the measured environment data to calculate the predicted driving speed of the robot; Step A5: The motor control module determines the predicted driving speed of the robot and controls the motor output through the PID algorithm so that the driving speed of the robot is consistent with the predicted driving speed; Step A6: If an obstacle is encountered during driving, the infrared obstacle avoidance module controls the robot to turn, avoid the obstacle, and search for the tennis ball again; Step A7: When the robot moves near the tennis ball, the robotic arm control module controls the robotic arm to pick up the tennis ball, put it into the box, increase the number by one, and determine whether the box is full. If not, continue to pick up tennis balls. If full, notify the user.

Images