Calf feeding robot with autonomous navigation control system and use method

Abstract

The invention provides a calf feeding robot with an autonomous navigation control system and a use method, and relates to the crossing field of robot technology and animal husbandry. A calf feeding robot with an autonomous navigation control system comprises a movable vehicle body chassis, a control screen, a main control bin, a first camera, a second camera, a laser radar base, a laser radar, a milk barrel, an electromagnetic valve, a nipple, a CAN communication interface module and a control panel. An autonomous navigation control system is arranged in the control panel and comprises an environment sensing subsystem, a navigation control subsystem, a milk feeding subsystem and a vehicle body chassis control subsystem, a highly-integrated control framework is achieved, and the integrated control system improves the operation efficiency and the response speed of the robot. Therefore, the robot can quickly adapt to the change of pasture environment and timely and accurately provide feeding service for calves.

Description

Technical Field

[0001] The present invention belongs to the intersection field of robotics and animal husbandry, and in particular relates to a calf feeding robot with an autonomous navigation control system and a use method thereof. Background Art

[0002] Animal husbandry is an important branch of agriculture, mainly involving the breeding and management of various domestic animals and livestock. It not only provides people with food and raw materials such as meat, dairy products, leather, etc., but also plays an important role in maintaining ecological balance and promoting rural economic development.

[0003] Calf feeding is a key link in animal husbandry, especially in dairy farming and beef cattle fattening. Correct feeding methods and management have a significant impact on the healthy growth, production performance and later economic benefits of calves. Calf feeding is not only related to the growth and health of individual animals, but also affects the efficiency and economic returns of the entire production system. Adopting scientific and reasonable feeding methods is of great significance to improving the benefits of animal husbandry.

[0004] In today's animal husbandry, calf feeding technology continues to develop, and a variety of modern means are used to improve feeding efficiency and management level. Among them, artificial feeding is still the main feeding method in many farms, and farm workers feed the calves at regular intervals according to the age and needs of the calves. However, the traditional artificial feeding method is costly, inefficient and risky. Some modern farms have also introduced automated feeding equipment, which can automatically provide milk according to the set time and the needs of the calves. However, the equipment and maintenance costs of automated and intelligent feeding systems are high, which are often difficult for small farms to afford. In addition, the current automatic feeding equipment does not have autonomous navigation functions, making it difficult to cope with complex pasture environments and posing certain safety risks. Therefore, developing a pasture autonomous navigation feeding robot is an important way to improve pasture management efficiency and reduce costs. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention proposes a calf feeding robot with an autonomous navigation control system and a method of use. The robot has the functions of autonomous navigation and intelligent feeding, providing accurate and quantitative feed for livestock, while reducing waste and improving feeding efficiency.

[0006] A calf feeding robot with an autonomous navigation control system includes a movable chassis, a control screen, a main control compartment, a first camera, a second camera, a laser radar base, a laser radar, a milk bucket, a solenoid valve, a nipple, a CAN communication interface module and a control board;

[0007] The movable chassis is a supporting structure for carrying a control screen, a main control compartment, a first camera, a second camera, a laser radar base, a laser radar, a milk bucket, a solenoid valve, a flow rate valve, a nipple, a CAN communication interface module and a control board, and provides a mobile function through a plurality of drive motors installed on the movable chassis; the plurality of drive motors are connected to the CAN communication interface module and communicate with the control board using the CAN communication protocol, and a power supply for power supply is provided on the movable chassis;

[0008] The control screen is placed outside the main control compartment and connected to the control board via an HDMI cable, and is used to display the image information collected by the first camera and the second camera through a graphical interface, while providing a human-computer interaction function to the user;

[0009] The main control compartment is fixed above the movable chassis, a control panel is fixedly installed inside, and a first camera, a second camera, a laser radar base and a laser radar are placed on the top;

[0010] The first camera and the second camera both provide infrared and LED lighting, and the first camera and the second camera are used to capture the image information of the left front and the right front respectively and send them to the control board;

[0011] The laser radar base is installed above the main control cabin and is used to fix the laser radar;

[0012] The laser radar is fixed by a laser radar support and is used to scan the surrounding environment, obtain the distance to the surrounding objects and the scanning angle of the laser and send them to the control board, and use the built-in IMU module to obtain the current coordinates and yaw angle and send them to the control board;

[0013] The milk bucket and the main control compartment are placed side by side on the movable chassis for storing milk;

[0014] The nipple is connected to the milk bucket via a solenoid valve and a flow rate valve and is used to feed the calf;

[0015] The solenoid valve is used to control whether the milk bucket and the nipple are connected;

[0016] The flow rate valve is used to control the flow rate of milk;

[0017] The CAN communication interface module is used to realize the communication between the movable chassis and the control panel through the CAN communication protocol;

[0018] The control panel is used to realize autonomous navigation according to user settings, image information collected by the first camera and the second camera, the distance to surrounding objects collected by the laser radar and the scanning angle of the laser, control the movement of the movable chassis, the switch of the solenoid valve and the flow rate of the flow rate valve, complete the calf feeding, and display the current coordinates and yaw angle on the remote control end;

[0019] The user settings include a travel speed set by the user and a planned path selected by the user;

[0020] The control board is equipped with an autonomous navigation control system, including an environment perception subsystem, a navigation control subsystem, a body chassis control subsystem and a feeding subsystem;

[0021] The environment perception subsystem is based on the Linux operating system and integrates the robot operating system. It generates point cloud data based on the distance to the surrounding objects measured by the laser radar and the scanning angle of the laser, processes the image information and the point cloud data, and sends the image information and the processed point cloud data to the navigation control subsystem.

[0022] The processing includes data format conversion, coordinate system conversion, filtering, downsampling and feature extraction;

[0023] The navigation control subsystem is based on Linux and the robot operating system, and is used to receive image information and processed point cloud data. When it is run for the first time or when the use environment changes, the map building program is used to process the image information into a two-dimensional grid map through the gmapping algorithm, and the processed point cloud data is converted into positioning information through the Robot Operating System open source framework using the positioning program. The position of the calf is determined according to the position of the calf in the two-dimensional grid map and the positioning information. According to the two-dimensional grid map and the position of the calf, the A* and DWA algorithms are used to perform global and local path planning respectively, and several complete planned paths are obtained and stored to ensure safety and reliability in complex environments;

[0024] In addition to changes in the operating and use environment, it is used to adjust the local planned path and avoid obstacles based on the planned path selected by the user according to the real-time acquired image information and point cloud data using the DWA algorithm, and send the adjusted local planned path to the body chassis control subsystem;

[0025] The chassis control subsystem is used to generate a drive signal according to the planned path and user settings, and send the drive signal to the drive motor of the movable chassis through the CAN communication interface module, so that the movable chassis moves according to the planned path;

[0026] The feeding subsystem is used to control the solenoid valve to be energized after reaching the position of the calf, and use the nipple to feed the milk in the milk bucket to the calf, and control the solenoid valve to be de-energized after the feeding is completed;

[0027] The graphical interface includes a chassis control area, an environment perception area and a navigation control area;

[0028] The chassis control area includes a path selection box, a speed input box, a chassis start button and a movement control area;

[0029] The path selection box is used for the user to select the planned path for this feeding task from a number of planned paths. After confirming the planned path, click the OK button, and the movable chassis will run according to the selected planned path to achieve the movement of the established planned path;

[0030] The speed input box is used for the user to input the travel speed of the movable chassis within a set range. After the input is completed, the user clicks the OK button to control the movable chassis to move at the set travel speed;

[0031] The chassis start button is pressed before path selection, speed input and movement control to start the drive motor of the movable chassis;

[0032] The mobile control area is used to control the movable chassis to move in four directions: front, back, left, and right;

[0033] The environment sensing area includes two camera start buttons, an image display area of ​​the first camera, and an image display area of ​​the second camera;

[0034] The two camera activation buttons are used to activate the first camera and the second camera;

[0035] The image display area of ​​the first camera and the image display area of ​​the second camera are respectively used to display image information collected in real time by the first camera and the second camera;

[0036] The navigation control area includes a navigation start button, a map save button and a map creation button;

[0037] The map creation button is used to start the map creation program and create a two-dimensional grid map;

[0038] The Save Map button is used to save the created two-dimensional grid map;

[0039] Navigation start button, used to start navigation and start traveling along the planned route;

[0040] A method for using a calf feeding robot with an autonomous navigation control system specifically comprises the following steps:

[0041] Step 1: Power on the movable chassis, click two camera start buttons on the remote control terminal or the graphical interface of the control screen to start the first camera and the second camera, the control board is connected to the drive motor on the movable chassis through the CAN communication interface module, click the chassis start button, and the control board transmits the generated drive signal to the drive motor through the CAN communication protocol;

[0042] Step 2: Click the map creation button on the graphical interface of the remote control terminal or the control screen to start the map creation program in the navigation control subsystem, enter the set speed in the speed input box on the graphical interface of the remote control terminal or the control screen, and use the mobile control area on the graphical interface of the remote control terminal to control the drive motor in the movable body chassis to move at the set speed and direction until a complete travel process is completed in the environment. During the travel process, the first camera and the second camera are used to collect image information of the surrounding environment, and the laser radar is used to measure the distance to the surrounding objects and the scanning angle of the laser, and the information is stored in the environment perception subsystem of the control panel;

[0043] Step 3: The environment perception subsystem generates point cloud data based on the distance to surrounding objects measured by the lidar and the scanning angle of the laser, processes the point cloud data, and sends the image information and the processed point cloud data to the navigation control subsystem;

[0044] Step 4: The map creation program processes the image information into a two-dimensional grid map through the gmapping algorithm. Click the Save Map button on the remote control terminal or the graphical interface of the control screen to store the generated two-dimensional grid map.

[0045] Step 5: The stored two-dimensional grid map is displayed on the control screen and the graphical interface of the remote control terminal, and the position of the calf in the two-dimensional grid map is set as the target point. The positioning program in the navigation control subsystem uses the Robot Operating System open source framework to convert the processed point cloud data into positioning information, and then obtains the positioning information of the target point, that is, the position of the calf;

[0046] Step 6: According to the two-dimensional grid map and the positioning information of the target point, the navigation control subsystem uses the A* algorithm and the DWA algorithm to perform global path planning and local path planning respectively, obtains several complete planned paths and stores them;

[0047] Step 7: Select the planned path for this calf feeding in the path selection box on the graphical interface of the remote control terminal or the control screen, click the navigation start button, and the chassis control subsystem generates a drive signal according to the selected planned path and user settings, and sends the drive signal to the drive motor of the movable chassis through the CAN communication interface module, so that the movable chassis starts to move according to the planned path;

[0048] Step 8: During the movement, the IMU module built into the laser radar obtains the current coordinates and yaw angle of the calf feeding robot with an autonomous navigation control system and displays them on the remote control terminal. At the same time, the navigation control subsystem uses the DWA algorithm to adjust the local planned path and achieve obstacle avoidance based on the real-time image information and point cloud data, and sends the adjusted local planned path to the chassis control subsystem, and sends the drive signal to the drive motor of the movable chassis through the CAN communication interface module to adjust and control the movement path of the movable chassis.

[0049] Step 9: When the movable chassis moves to the target point, it stops moving. The feeding subsystem controls the solenoid valve to be energized, the nipple is connected to the milk bucket, the feeding amount of the day is predicted, and the flow rate valve is used to control the milk flow rate according to the predicted feeding amount. The calf stretches its neck and uses the nipple to drink milk autonomously. After the set time, the solenoid valve is de-energized.

[0050] The method for controlling the milk flow rate by using a flow rate valve is as follows:

[0051] S1: Obtain the daily feeding amount record within a period of history;

[0052] S2: Store the daily feeding amount records in an array or list, and create an array of days corresponding to the feeding amount records, where the elements represent the days corresponding to the feeding amount records;

[0053] S3: Initialize a linear regression model, input the number of days as input features into the linear regression model, use the daily feeding amount record as the target value, train the linear regression model, and obtain a trained linear regression model;

[0054] S4: determining a feeding date, and determining the day corresponding to the date in the day array, and inputting the day into a linear regression model to obtain a predicted feeding amount;

[0055] S5: Calculate and set the milk flow rate based on the predicted feeding amount and the set feeding time to ensure that the appropriate amount of milk is provided within the scheduled time;

[0056] The predicted feeding amount and the set milk flow rate are displayed on a graphical interface;

[0057] S6: obtaining an actual feeding amount and comparing it with the predicted feeding amount, and if the difference between the two is greater than a set threshold, adjusting the parameters of the linear regression model according to the actual feeding amount to improve future predictions;

[0058] Step 10: The movable chassis moves to the next target point to continue feeding until the feeding task is completed. The calf feeding robot with an autonomous navigation control system returns to the starting point and waits for the next round of feeding task or charging.

[0059] Compared with the prior art, the beneficial effects of the present invention are:

[0060] The present invention proposes a calf-feeding robot with an autonomous navigation control system, wherein the autonomous navigation control system carried by the control panel includes an environmental perception subsystem, a navigation control subsystem, a feeding subsystem and a body chassis control subsystem, realizing a highly integrated control architecture. This integrated control system improves the operating efficiency and response speed of the robot, enabling the robot to quickly adapt to changes in the pasture environment and provide feeding services for calves in a timely and accurate manner.

[0061] The environmental perception subsystem uses multi-sensor fusion technology to achieve comprehensive perception of the ranch environment, including obstacle detection and terrain recognition, which improves the robot's perception accuracy of the environment, reduces misjudgments and missed judgments, thereby enhancing the robot's autonomous navigation capabilities in complex ranch environments and ensuring the safety of the robot and calves.

[0062] The navigation and control subsystem uses the A* and DWA algorithms for global and local path planning. It can calculate the optimal path in real time based on the data provided by the environmental perception subsystem, allowing the robot to choose the shortest or safest path when performing tasks, thereby improving the efficiency of the feeding task, reducing energy consumption, and extending the robot's battery life.

[0063] The feeding subsystem can automatically adjust the feeding amount and time according to the calf's habits and needs to achieve personalized feeding. Personalized feeding not only meets the nutritional needs of different calves, but also improves the calf's healthy growth rate, while reducing milk waste and improving the economic benefits of the ranch;

[0064] The chassis control subsystem achieves smooth movement and precise positioning of the robot through precise drive control. Precise drive control ensures the stability of the robot during movement, reduces interference with the calf, and improves the robot's ability to pass through complex terrain, ensuring the smooth progress of the feeding task. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 Schematic diagram of the structure of a calf feeding robot with an autonomous navigation control system in an embodiment of the present invention;

[0066] Figure 2 It is a partial structural schematic diagram of a calf feeding robot with an autonomous navigation control system in an embodiment of the present invention;

[0067] Figure 3 It is a structural block diagram of an autonomous navigation control system of a calf feeding robot with an autonomous navigation control system in an embodiment of the present invention;

[0068] Figure 4 This is a schematic diagram of a graphical interface of a calf feeding robot with an autonomous navigation control system according to an embodiment of the present invention;

[0069] Among them, 1- movable chassis, 2- control screen, 3- main control compartment, 4- first camera, 5- second camera, 6- lidar base, 7- lidar, 8- milk bucket, 9- nipple, 10- can communication interface module, 11- control panel, 12- path selection box, 13- speed input box, 14- chassis start button, 15- camera start button, 16- navigation start button, 17- save map button, 18- map creation button, 19- mobile control area. DETAILED DESCRIPTION

[0070] The present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0071] like Figure 1 and Figure 2 As shown, a calf feeding robot with an autonomous navigation control system includes a movable chassis 1, a control screen 2, a main control compartment 3, a first camera 4, a second camera 5, a laser radar base 6, a laser radar 7, a milk bucket 8, a solenoid valve, a nipple 9, a CAN communication interface module 10 and a control board 11;

[0072] The movable chassis 1 is a supporting structure, used to carry the control screen 2, the main control compartment 3, the first camera 4, the second camera 5, the laser radar base 6, the laser radar 7, the milk bucket 8, the solenoid valve, the flow rate valve, the nipple 9, the CAN communication interface module 10 and the control board 11, and provides a mobile function through a number of drive motors installed on the movable chassis 1; the chassis is a front-to-rear drive Ackerman structure, and the overall bridge suspension design ensures stability under complex terrain. The several drive motors are connected to the CAN communication interface module 10 and communicate with the control board 11 using the CAN communication protocol. At the same time, a power supply for power supply is provided on the movable chassis 1;

[0073] The maximum load of the movable chassis 1 in this embodiment is 100 kg;

[0074] The control screen 2 is placed outside the main control compartment 3 and connected to the control board 11 via an HDMI cable, and is used to display the image information collected by the first camera 4 and the second camera 5 through a graphical interface, while providing a human-computer interaction function to the user for easy operation and monitoring;

[0075] The control screen 2 in this embodiment is a 1270*720 touch screen that can perform 5-point capacitive touch;

[0076] The main control compartment 3 is fixed above the movable chassis 1, a control panel 11 is fixedly installed inside, and a first camera 4, a second camera 5, a laser radar base 6 and a laser radar 7 are placed on the top;

[0077] The first camera 4 and the second camera 5 provide infrared and LED lighting to ensure visual perception in low light and night conditions. The first camera 4 and the second camera 5 are used to capture the image information of the left front and the right front respectively and send them to the control board 11 to provide intuitive visual feedback for the operator;

[0078] In this embodiment, the first camera 4 and the second camera 5 are arranged on both sides of the front of the vehicle to form a 45° viewing angle. The first camera 4 and the second camera 5 can perform 120-degree ultra-wide-angle photography;

[0079] The laser radar base 6 is installed above the main control cabin 3 and is used to fix the laser radar 7;

[0080] The laser radar 7 is fixed by the laser radar support 6, and is used to scan the surrounding environment, obtain the distance to the surrounding objects and the scanning angle of the laser, and send them to the control board 11, so as to assist the robot in accurate positioning and navigation. Considering that the robot may encounter uneven road surface during the journey, in serious cases, it will cause the robot to deviate and fail to reach the designated point. To avoid this situation, the built-in IMU module is used to obtain the current coordinates and yaw angle and send them to the control board 11;

[0081] The built-in IMU module in this embodiment includes a 3-axis accelerometer and a 3-axis gyroscope;

[0082] In this embodiment, the laser radar 7 can detect objects at a minimum distance of 0.1 meters and a maximum distance of 100 meters, and the point cloud frame rate is 10Hz;

[0083] The milk bucket 8 and the main control compartment 3 are placed side by side on the movable chassis 1 for storing milk;

[0084] The nipple 9 is connected to the milk bucket through a solenoid valve and a flow rate valve, and is used to feed the calf;

[0085] In this embodiment, the nipple 9 is a common livestock drinking water device on the market. When no external force is applied, the liquid will not flow out on its own. The nipple 9 is a sucking type, a direct discharge type or other structural types to meet the requirements of users in different scenarios.

[0086] The solenoid valve is used to control whether the milk bucket 8 is connected to the nipple 9;

[0087] The flow rate valve is used to control the flow rate of milk;

[0088] The CAN communication interface module 10 is used to realize the communication between the movable chassis 1 and the control board 11 through the CAN communication protocol;

[0089] The control panel 11 is used to realize autonomous navigation, control the movement of the movable chassis 1, the switch of the solenoid valve and the flow rate of the flow rate valve according to the user settings, the image information collected by the first camera 4 and the second camera 5, the distance to the surrounding objects collected by the laser radar 7 and the scanning angle of the laser, complete the calf feeding, and display the current coordinates and yaw angle on the remote control end; the user settings include the travel speed set by the user and the planned path selected by the user;

[0090] The control panel 11 is equipped with an autonomous navigation control system. Figure 3 As shown, it includes an environment perception subsystem, a navigation control subsystem, a body chassis control subsystem and a feeding subsystem;

[0091] The environment perception subsystem is based on the Linux operating system and integrates the Robot Operating System (ROS). It generates point cloud data based on the distance to surrounding objects measured by the laser radar 7 and the scanning angle of the laser, processes the image information and the point cloud data, and sends the image information and the processed point cloud data to the navigation control subsystem to provide basic data for subsequent path planning; the processing includes data format conversion, coordinate system conversion, filtering, downsampling and feature extraction;

[0092] In this embodiment, the raw point cloud data generated by the laser radar 7 may need to be converted into a format suitable for processing. For example, the point cloud data may need to be converted from the specific format of the laser radar 7 to a universal point cloud format, such as PCD (Point Cloud Data) or PLY; point cloud data are usually generated in the coordinate system of the laser radar 7, but in order to perform path planning, these data need to be converted to a global coordinate system or a robot coordinate system, which involves the use of a transformation matrix for rotation and translation; the raw point cloud data may contain noise and unnecessary points, so it may be necessary to remove outliers through a filter and reduce the number of points through downsampling for ease of processing and analysis; useful features such as edges, corners and planes are extracted from the point cloud data, which are crucial for subsequent path planning and obstacle detection;

[0093] The navigation control subsystem is based on Linux and the robot operating system, and is used to receive image information and processed point cloud data. When it is run for the first time or when the use environment changes, the map building program is used to process the image information into a two-dimensional grid map through the gmapping algorithm, and the processed point cloud data is converted into positioning information through the Robot Operating System open source framework using the positioning program. The position of the calf is determined according to the position of the calf in the two-dimensional grid map and the positioning information. According to the two-dimensional grid map and the position of the calf, the A* and DWA algorithms are used to perform global and local path planning respectively, and several complete planned paths are obtained and stored to ensure safety and reliability in complex environments;

[0094] In addition to changes in the operating and use environment, it is used to adjust the local planned path and avoid obstacles based on the planned path selected by the user according to the real-time acquired image information and point cloud data using the DWA algorithm, and send the adjusted local planned path to the body chassis control subsystem;

[0095] The chassis control subsystem is used to generate a driving signal according to the planned path and user settings, and send the driving signal to the driving motor of the movable chassis 1 through the CAN communication interface module 10, so that the movable chassis 1 moves according to the planned path;

[0096] The feeding subsystem is used to control the solenoid valve to be energized after reaching the position of the calf, and use the nipple 9 to feed the milk in the milk bucket 8 to the calf, and control the solenoid valve to be de-energized after the feeding is completed;

[0097] like Figure 4 As shown, the graphical interface includes a chassis control area, an environment perception area and a navigation control area;

[0098] The chassis control area includes a path selection box 12, a speed input box 13, a chassis start button 14 and a movement control area 19;

[0099] The path selection box 12 is used for the user to select a planned path for this feeding task from a number of planned paths. After confirming the planned path, click the OK button, and the movable chassis 1 can run according to the selected planned path to achieve the movement of the established planned path;

[0100] The speed input box 13 is used for the user to input the travel speed of the movable chassis 1 within a set range. After the input is completed, the user clicks the OK button to control the movable chassis 1 to move at the set travel speed;

[0101] The speed setting range in this embodiment is 0m / s-2m / s;

[0102] The chassis start button 14 is pressed before path selection, speed input and movement control, and is used to start the drive motor of the movable body chassis 1;

[0103] The movement control area 19 is used to control the movable chassis 1 to move in four directions: front, back, left, and right;

[0104] The environment sensing area includes two camera start buttons 15, an image display area of ​​the first camera 4, and an image display area of ​​the second camera 5;

[0105] The two camera activation buttons 15 are used to activate the first camera 4 and the second camera 5;

[0106] The image display area of ​​the first camera 4 and the image display area of ​​the second camera 5 are respectively used to display image information collected in real time by the first camera 4 and the second camera 5;

[0107] The navigation control area includes a navigation start button 16, a save map button 17 and a map creation button 18;

[0108] The map creation button 18 is used to start the map creation program and create a two-dimensional grid map;

[0109] The save map button 17 is used to save the created two-dimensional grid map;

[0110] The navigation start button 16 is used to start navigation and start traveling along the planned route;

[0111] A method for using a calf feeding robot with an autonomous navigation control system specifically comprises the following steps:

[0112] Step 1: Power on the movable chassis 1, click two camera start buttons 15 on the remote control terminal or the graphical interface of the control screen 2 to start the first camera 4 and the second camera 5, the control board 11 is connected to the drive motor on the movable chassis 1 through the CAN communication interface module 10, click the chassis start button 14, and the control board 11 transmits the generated drive signal to the drive motor through the CAN communication protocol;

[0113] Step 2: Click the map creation button 18 on the graphical interface of the remote control terminal or the control screen 2 to start the map creation program in the navigation control subsystem, enter the set speed in the speed input box 13 on the graphical interface of the remote control terminal or the control screen 2, and use the mobile control area 19 on the graphical interface of the remote control terminal to control the drive motor in the movable body chassis 1 to move at the set speed and direction until a complete travel process is completed in the environment. During the travel process, the first camera 4 and the second camera 5 are used to collect image information of the surrounding environment, and the laser radar 7 is used to measure the distance to the surrounding objects and the scanning angle of the laser, and the information is stored in the environment perception subsystem of the control board 11;

[0114] Step 3: The environment perception subsystem generates point cloud data based on the distance to surrounding objects measured by the laser radar 7 and the scanning angle of the laser, processes the point cloud data, and sends the image information and the processed point cloud data to the navigation control subsystem;

[0115] Step 4: The map creation program processes the image information into a two-dimensional grid map through the gmapping algorithm, and clicks the save map button 17 on the graphical interface of the remote control terminal or the control screen 2 to store the generated two-dimensional grid map;

[0116] Step 5: The stored two-dimensional grid map is displayed on the control screen 2 and the graphical interface of the remote control terminal, and the position of the calf in the two-dimensional grid map is set as the target point. The positioning program in the navigation control subsystem uses the Robot Operating System open source framework to convert the processed point cloud data into positioning information, and then obtains the positioning information of the target point, that is, the position of the calf;

[0117] Step 6: According to the two-dimensional grid map and the positioning information of the target point, the navigation control subsystem uses the A* algorithm and the DWA algorithm to perform global path planning and local path planning respectively, obtains several complete planned paths and stores them;

[0118] Step 7: Select the planned path for feeding the calf in the path selection box 12 on the graphical interface of the remote control terminal or the control screen 2, click the navigation start button 16, the chassis control subsystem generates a drive signal according to the selected planned path and user settings, and sends the drive signal to the drive motor of the movable chassis 1 through the CAN communication interface module 10, so that the movable chassis 1 starts to move along the planned path;

[0119] Step 8: During the travel, the IMU module built into the laser radar 7 obtains the current coordinates and yaw angle of the calf feeding robot with an autonomous navigation control system and displays them on the remote control terminal. At the same time, the navigation control subsystem uses the DWA algorithm to adjust the local planned path and avoid obstacles based on the real-time image information and point cloud data, and sends the adjusted local planned path to the chassis control subsystem, and sends the drive signal to the drive motor of the movable chassis 1 through the CAN communication interface module 10 to adjust and control the travel path of the movable chassis 1;

[0120] Step 9: When the movable chassis 1 moves to the target point and stops, the feeding subsystem controls the electromagnetic valve to be energized, the nipple 9 is connected to the milk bucket 8, the feeding amount of the day is predicted, and the flow rate valve is used to control the milk flow rate according to the predicted feeding amount. The calf stretches its neck and drinks milk autonomously using the nipple 9. After the set time, the electromagnetic valve is de-energized;

[0121] The method for controlling the milk flow rate by using a flow rate valve is as follows:

[0122] S1: Obtain the daily feeding amount record within a period of history;

[0123] In this embodiment, the daily feeding volume record in the past week is obtained, for example: [2.5, 2.6, 2.7, 2.6, 2.5, 2.4, 2.5] liters;

[0124] S2: Store the daily milk volume records in an array or list, such as milk_volumes[], and create an array of days corresponding to the milk volume records, such as days[], where the elements represent the days corresponding to the milk volume records;

[0125] S3: Initialize a linear regression model, input the number of days as input features into the linear regression model, use the daily feeding amount record as the target value, train the linear regression model, and obtain a trained linear regression model;

[0126] S4: determining a feeding date, and determining the day corresponding to the date in the day array, and inputting the day into a linear regression model to obtain a predicted feeding amount;

[0127] S5: Calculate and set the milk flow rate based on the predicted feeding amount and the set feeding time to ensure that the appropriate amount of milk is provided within the scheduled time;

[0128] The predicted feeding amount and the set milk flow rate are displayed on a graphical interface;

[0129] S6: obtaining an actual feeding amount and comparing it with the predicted feeding amount, and if the difference between the two is greater than a set threshold, adjusting the parameters of the linear regression model according to the actual feeding amount to improve future predictions;

[0130] Step 10: The movable chassis 1 moves to the next target point to continue feeding until the feeding task is completed. The calf feeding robot with an autonomous navigation control system returns to the starting point and waits for the next round of feeding task or charging.

Claims

1. A calf feeding robot with an autonomous navigation control system, characterized in that: It includes a movable chassis, a control screen, a main control compartment, a first camera, a second camera, a laser radar base, a laser radar, a milk bucket, a solenoid valve, a nipple, a CAN communication interface module and a control panel; The movable chassis is a supporting structure for carrying a control screen, a main control compartment, a first camera, a second camera, a laser radar base, a laser radar, a milk bucket, a solenoid valve, a flow rate valve, a nipple, a CAN communication interface module and a control board, and provides a mobile function through a plurality of drive motors installed on the movable chassis; the plurality of drive motors are connected to the CAN communication interface module and communicate with the control board using the CAN communication protocol, and a power supply for power supply is provided on the movable chassis; The control screen is placed outside the main control compartment and connected to the control board via an HDMI cable, and is used to display the image information collected by the first camera and the second camera through a graphical interface, while providing a human-computer interaction function to the user; The main control compartment is fixed above the chassis of the vehicle body, a control panel is fixedly installed inside, and a first camera, a second camera, a laser radar base and a laser radar are placed on the top; The first camera and the second camera both provide infrared and LED lighting, and the first camera and the second camera are used to capture the image information of the left front and the right front respectively and send them to the control board; The laser radar base is installed above the main control cabin and is used to fix the laser radar; The laser radar is fixed by a laser radar support and is used to scan the surrounding environment, obtain the distance to the surrounding objects and the scanning angle of the laser and send them to the control board, and use the built-in IMU module to obtain the current coordinates and yaw angle and send them to the control board; The milk bucket is used to store milk; The nipple is connected to the milk bucket via a solenoid valve and a flow rate valve and is used to feed the calf; The solenoid valve is used to control whether the milk bucket and the nipple are connected; The flow rate valve is used to control the flow rate of milk; The CAN communication interface module is used to realize the communication between the vehicle chassis and the control board through the CAN communication protocol; The control panel is used to realize autonomous navigation according to user settings, image information collected by the first camera and the second camera, the distance to surrounding objects collected by the lidar and the scanning angle of the laser, control the movement of the vehicle chassis, the switching of the solenoid valve and the flow of the flow rate valve, complete the calf feeding, and display the current coordinates and yaw angle on the remote control terminal.

2. A calf feeding robot with an autonomous navigation control system according to claim 1, characterized in that: The user settings include a travel speed set by the user and a planned path selected by the user.

3. The calf feeding robot with an autonomous navigation control system according to claim 1, characterized in that: The control board is equipped with an autonomous navigation control system, including an environment perception subsystem, a navigation control subsystem, a body chassis control subsystem and a feeding subsystem; The environment perception subsystem is based on the Linux operating system and integrates the robot operating system. It generates point cloud data based on the distance to the surrounding objects measured by the laser radar and the scanning angle of the laser, processes the image information and the point cloud data, and sends the image information and the processed point cloud data to the navigation control subsystem. The processing includes data format conversion, coordinate system conversion, filtering, downsampling and feature extraction; The navigation control subsystem is based on Linux and the robot operating system, and is used to receive image information and processed point cloud data. When it is run for the first time or when the use environment changes, the map building program is used to process the image information into a two-dimensional grid map through the gmapping algorithm, and the processed point cloud data is converted into positioning information through the Robot Operating System open source framework using the positioning program. The position of the calf is determined according to the position of the calf in the two-dimensional grid map and the positioning information. According to the two-dimensional grid map and the position of the calf, the A* and DWA algorithms are used to perform global and local path planning respectively, and several complete planned paths are obtained and stored to ensure safety and reliability in complex environments; In addition to changes in the operating and use environment, it is used to adjust the local planned path and avoid obstacles based on the planned path selected by the user according to the real-time acquired image information and point cloud data using the DWA algorithm, and send the adjusted local planned path to the body chassis control subsystem; The chassis control subsystem is used to generate a drive signal according to the planned path and user settings, and send the drive signal to the drive motor of the chassis through the CAN communication interface module, so that the chassis moves according to the planned path; The feeding subsystem is used to control the solenoid valve to be energized after reaching the position of the calf, and use the nipple to feed the milk in the milk bucket to the calf, and control the solenoid valve to be de-energized after the feeding is completed.

4. The calf feeding robot with an autonomous navigation control system according to claim 1, characterized in that: The graphical interface includes a chassis control area, an environment perception area and a navigation control area; The chassis control area includes a path selection box, a speed input box, a chassis start button and a movement control area; The path selection box is used for the user to select a planned path for this feeding task from a number of planned paths. After confirming the planned path, click the OK button, and the chassis will run according to the selected planned path to achieve the movement of the established planned path; The speed input box is used for the user to input the speed of the chassis within a set range. After the input is completed, click the OK button to control the chassis to move at the set speed. The chassis start button is pressed before path selection, speed input and movement control to start the driving motor of the chassis; The movement control area is used to control the vehicle chassis to move in four directions: front, back, left, and right; The environment sensing area includes two camera start buttons, an image display area of ​​the first camera, and an image display area of ​​the second camera; The two camera activation buttons are used to activate the first camera and the second camera; The image display area of ​​the first camera and the image display area of ​​the second camera are respectively used to display image information collected in real time by the first camera and the second camera; The navigation control area includes a navigation start button, a map save button and a map creation button; The map creation button is used to start the map creation program and create a two-dimensional grid map; The Save Map button is used to save the created two-dimensional grid map; The navigation start button is used to start navigation and start moving along the planned route.

5. The method for using the calf feeding robot with an autonomous navigation control system according to claim 1, characterized in that: The following steps are involved: Step 1: Power on the movable chassis, click two camera start buttons on the remote control terminal or the graphical interface of the control screen to start the first camera and the second camera, the control board is connected to the drive motor on the movable chassis through the CAN communication interface module, click the chassis start button, and the control board transmits the generated drive signal to the drive motor through the CAN communication protocol; Step 2: Click the map creation button on the graphical interface of the remote control terminal or the control screen to start the map creation program in the navigation control subsystem, enter the set speed in the speed input box on the graphical interface of the remote control terminal or the control screen, and use the mobile control area on the graphical interface of the remote control terminal to control the drive motor in the chassis to move at the set speed and direction until a complete travel process is completed in the environment. During the travel process, the first camera and the second camera are used to collect image information of the surrounding environment, and the laser radar is used to measure the distance to the surrounding objects and the scanning angle of the laser, and the information is stored in the environment perception subsystem of the control panel; Step 3: The environment perception subsystem generates point cloud data based on the distance to surrounding objects measured by the lidar and the scanning angle of the laser, processes the point cloud data, and sends the image information and the processed point cloud data to the navigation control subsystem; Step 4: The map creation program processes the image information into a two-dimensional grid map through the gmapping algorithm. Click the Save Map button on the remote control terminal or the graphical interface of the control screen to store the generated two-dimensional grid map. Step 5: The stored two-dimensional grid map is displayed on the control screen and the graphical interface of the remote control terminal, and the position of the calf in the two-dimensional grid map is set as the target point. The positioning program in the navigation control subsystem uses the Robot Operating System open source framework to convert the processed point cloud data into positioning information, and then obtains the positioning information of the target point, that is, the position of the calf; Step 6: According to the two-dimensional grid map and the positioning information of the target point, the navigation control subsystem uses the A* algorithm and the DWA algorithm to perform global path planning and local path planning respectively, obtains several complete planned paths and stores them; Step 7: Select the planned path for this calf feeding in the path selection box on the graphical interface of the remote control terminal or the control screen, and click the navigation start button. The chassis control subsystem generates a drive signal according to the selected planned path and user settings, and sends the drive signal to the drive motor of the chassis through the CAN communication interface module, so that the chassis starts to move along the planned path; Step 8: During the movement, the IMU module built into the laser radar obtains the current coordinates and yaw angle of the calf feeding robot with an autonomous navigation control system and displays them on the remote control terminal. At the same time, the navigation control subsystem uses the DWA algorithm to adjust the local planned path and achieve obstacle avoidance based on the real-time image information and point cloud data, and sends the adjusted local planned path to the chassis control subsystem, and sends the drive signal to the drive motor of the chassis through the CAN communication interface module to adjust and control the movement path of the chassis. Step 9: When the movable chassis moves to the target point, it stops moving. The feeding subsystem controls the solenoid valve to be energized, the nipple is connected to the milk bucket, the feeding amount of the day is predicted, and the flow rate valve is used to control the milk flow rate according to the predicted feeding amount. The calf stretches its neck and uses the nipple to drink milk autonomously. After the set time, the solenoid valve is de-energized. Step 10: The movable chassis moves to the next target point to continue feeding until the feeding task is completed. The calf feeding robot with an autonomous navigation control system returns to the starting point and waits for the next round of feeding task or charging.

6. The method for using the calf feeding robot with an autonomous navigation control system according to claim 5, characterized in that: The method for controlling the milk flow rate by using the flow rate valve in step 9 is: S1: Obtain the daily feeding amount record within a period of history; S2: Store the daily feeding amount records in an array or list, and create an array of days corresponding to the feeding amount records, where the elements represent the days corresponding to the feeding amount records; S3: Initialize a linear regression model, input the number of days as input features into the linear regression model, use the daily feeding amount record as the target value, train the linear regression model, and obtain a trained linear regression model; S4: determining a feeding date, and determining the day corresponding to the date in the day array, and inputting the day into a linear regression model to obtain a predicted feeding amount; S5: Calculate and set the milk flow rate based on the predicted feeding amount and the set feeding time to ensure that the appropriate amount of milk is provided within the scheduled time; The predicted feeding amount and the set milk flow rate are displayed on a graphical interface; S6: Obtain the actual feeding amount and compare it with the predicted feeding amount. If the difference between the two is greater than a set threshold, adjust the parameters of the linear regression model according to the actual feeding amount to improve future predictions.

Images