Feeding method and system based on intelligent sheep feeding robot

Abstract

The invention provides a feeding method and system based on an intelligent sheep feeding robot, and belongs to the technical field of feeding robos.The feeding method comprises the steps that according to a sheep feed feeding amount database and the number of sheep in different life cycle stages of each sheep pen, the feeding task and the stock bin material taking amount of each sheep pen are determined; determining feeding time and generating a task process based on the feeding task of each sheepfold and the material taking amount of the stock bin; scanning a sheep field area through a laser radar to generate a three-dimensional digital map; and determining the positions of a stock bin and each sheepfold based on the three-dimensional digital map, taking the stock bin as a starting point and an ending point, taking each sheepfold as a path point location, carrying out feeding path planning by taking the shortest path as a target, and controlling the robot to automatically execute a task. According to the method, automatic feed taking, path planning and accurate feeding can be achieved, farmers are helped to improve the feed feeding efficiency of the sheep, and the labor intensity of the farmers is relieved.

Description

Technical Field

[0001] The present invention relates to the technical field of feeding robots, and in particular to a feeding method, system and medium based on an intelligent sheep feeding robot. Background Art

[0002] With the transformation of the livestock production model from extensive to intensive development, the requirements for production efficiency are constantly increasing. However, problems such as low labor productivity and labor shortage have become bottlenecks restricting the rapid development of animal husbandry. In rural areas, animal husbandry is one of the important economic sources for farmers. However, traditional artificial feeding methods have many shortcomings, such as low feeding efficiency, serious feed waste, and high labor costs. In addition, with the reduction of rural labor force and the aggravation of aging problems, the difficulty and cost of artificial feeding have further increased. Therefore, the development of AI sheep feeding intelligent robots to realize automated and intelligent feeding management is of great significance to improving rural breeding efficiency and reducing labor costs.

[0003] Most existing feeding robots can only feed according to preset feeding plans. They lack flexible feeding strategies for sheep at different growth stages and of different breeds, cannot meet the needs of refined farming, and have a poor degree of automation. Summary of the Invention

[0004] To solve the above problems, the present invention provides a feeding method based on an intelligent sheep feeding robot, which can realize automatic material collection, path planning and precise feeding, helping farmers improve the feeding efficiency of sheep and reduce the labor intensity of farmers.

[0005] To achieve the above objectives, the present invention provides the following technical solutions.

[0006] A feeding system based on an intelligent sheep feeding robot comprises the following steps: Determine the feeding tasks and feed silo intake for each sheep pen based on the sheep feed intake database and the number of sheep at different life cycle stages in each sheep pen; Determine feeding time and generate task flow based on feeding tasks of each sheep pen and feed silo intake; Scan the sheep farm area with LiDAR to generate a three-dimensional digital map; The location of the feed silo and each sheep pen is determined based on a three-dimensional digital map. The feed silo is used as the starting and ending point, and each sheep pen is used as a path point. The feeding route is planned with the shortest path as the goal. According to the dynamic algorithm, the current vehicle-loaded material weight is dynamically calculated based on the material quantity taken from the silo and the speed of the unloading motor and the real-time weighing data of the robot scale; Wait for the robot to actively request feeding tasks, obtain feeding tasks and feed silo feed quantity, obtain task process, determine the silo, sheep house, and feeding route based on the feeding path planning results, determine the current machine vehicle load weight based on the real-time weighing data of the robot scale, and control the robot to automatically perform the task.

[0007] Preferably, the process of scanning the sheep farm area by laser radar to generate a three-dimensional digital map comprises the following steps: Use a laser radar scanning device to scan the sheep farm area with a laser beam and record the reflected signal to obtain point cloud data; Filter the point cloud data to obtain denoised point cloud data; Perform plane fitting on the point cloud data using the RANSAC algorithm to identify the structures of the ground, walls, and railings, and remove non-ground or non-important point cloud data; Based on the geographic information of the sheep farm, a three-dimensional data set is generated by aggregating the processed point cloud data according to the spatial coordinate system, so that each point cloud data has the corresponding spatial coordinates and additional information of color and intensity; Based on the 3D data set, the point cloud data is converted into mesh, surface or volume representation to construct a complete 3D digital map.

[0008] Preferably, the construction of the sheep feed amount database comprises the following steps: Identify the life cycle stages of sheep, including the lamb stage, young lamb stage and adult lamb stage; the lamb stage is from birth to 6 months; the young lamb stage is from 6 months to 12 months; and the adult lamb stage is over 12 months; Determine the feeding amount required for each sheep at different weights in different life cycle stages and establish a sheep feed feeding amount database.

[0009] Preferably, the method of determining the feeding task and feed silo withdrawal amount of each sheep pen according to the sheep feed feeding amount database and the number of sheep at different life cycle stages in each sheep pen comprises the following steps: Configure surveillance cameras to regularly collect surveillance images of each sheep pen; perform pre-processing on surveillance images by denoising, cropping, and standardization; Using computer vision technology, a pre-trained convolutional neural network model is used to identify pre-processed surveillance images. Based on the sheep's body shape and color characteristics, the number of sheep and their corresponding life cycle stage are determined. The weight of sheep is obtained through ground pressure sensors to verify the number of sheep at different life cycle stages; Based on the sheep feed feeding amount database and the number of sheep in different life cycle stages in each sheep pen, the feeding task and feed silo intake of each sheep pen are determined.

[0010] Preferably, the method of determining the positions of the feed silo and each sheep pen based on the three-dimensional digital map, taking the feed silo as the starting point and the end point, and each sheep pen as the path point, and planning the feeding path with the shortest path as the goal includes the following steps: Based on the three-dimensional digital map, mark the location of the feed silos and sheep pens within the sheep farm area; Convert the coordinate information in the three-dimensional digital map into a two-dimensional coordinate system; Based on the two-dimensional coordinate system, considering obstacles and site restrictions, the Dijkstra algorithm is used to calculate the shortest planar path from the silo to each sheep pen.

[0011] Based on the location and number of sheep pens, a greedy algorithm is used to optimize the order of visiting the sheep pens to ensure the shortest path.

[0012] Preferably, when the sheep feeding intelligent robot is controlled to take materials from the silo and feed the sheep in each sheep pen along the planned feeding path, the amount of material taken and the amount of feeding are determined by the pressure sensor data of the storage area of ​​the sheep feeding intelligent robot.

[0013] A feeding system based on an intelligent sheep feeding robot, the system comprising: The data acquisition module is used to obtain the three-dimensional point cloud data of the sheep farm area, collect monitoring images of each sheep pen in the sheep farm area, and the weight data of the sheep in the sheep pen; The Raspberry Pi controller is used to construct a 3D digital map of the sheep farm based on 3D point cloud data. It is also used to establish a database of sheep feed requirements at different life cycle stages. Monitoring images are used to identify the number of sheep at different life cycle stages. Based on the sheep feed requirement database, the feed silo intake and feeding tasks for each pen are determined based on the number and weight data of sheep at different life cycle stages in each pen. The system is also used to determine the location of the silo and each pen based on the 3D digital map, using the silo as the starting and end point and each pen as a path point to plan a feeding route with the shortest path as the goal. A dynamic algorithm is also used to dynamically calculate the current vehicle-loaded material weight based on the feed silo intake and the real-time weighing data from the robot scale. The robot is used to obtain feeding tasks and feed silo feed quantity, obtain task processes, determine the silo, sheep house, and feeding route based on the feeding path planning results, determine the current machine vehicle load weight based on the real-time weighing data of the robot scale, and control the robot to automatically perform tasks.

[0014] Preferably, it also includes: The cloud server is used to store the amount of feed taken from the silo, the feeding tasks of each sheep pen, and the feeding path planning results, and conduct data analysis to determine the feeding strategy; Mobile apps and farmers' PCs are used to obtain feed silo intake, feeding tasks for each sheep pen, feeding route planning results, and feeding strategies stored on the cloud server through wireless communication.

[0015] Beneficial effects of the present invention: The present invention proposes a feeding method based on an intelligent sheep feeding robot. The method obtains millimeter-level three-dimensional point cloud data of the sheep farm area through laser scanning, accurately reproduces the spatial distribution of terrain, buildings and obstacles, and builds a real environment base for path planning; constructs a three-dimensional digital map based on the point cloud classification algorithm, automatically distinguishes functional areas such as silos and sheep pens, generates a queryable semantic space index, and significantly improves the efficiency of path planning decisions. A feed feeding amount database is established to dynamically match the nutritional needs of sheep in different life cycles. The present invention integrates monitoring images and weight sensor data, uses a deep learning algorithm to realize the recognition of the sheep life cycle, and dynamically adjusts the feeding amount in combination with real-time weight data, effectively reducing feed waste. The three-dimensional path planning of this method realizes conflict-free path allocation and task scheduling, providing an intelligent and precise feeding solution for large-scale sheep farms. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a flow chart of a method according to an embodiment of the present invention; Figure 2 It is a system architecture diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0018] Example 1 This embodiment proposes a feeding method based on a sheep feeding intelligent robot, and the specific steps are as follows: Figure 1 As shown, including: S1: Determine the feeding task and feed silo takeout amount for each sheep pen based on the sheep feed feeding amount database and the number of sheep at different life cycle stages in each sheep pen.

[0019] S2: Based on the feeding tasks of each sheep pen and the feed intake of the feed silo, the feeding time is determined and the task flow is generated.

[0020] S3: Scan the sheep farm area with LiDAR to generate a three-dimensional digital map.

[0021] S4: Determine the location of the feed silo and each sheep pen based on the three-dimensional digital map, use the feed silo as the starting point and end point, and each sheep pen as the path point, and plan the feeding path with the shortest path as the goal.

[0022] S5: According to the dynamic algorithm, the current vehicle-loaded material weight is dynamically calculated based on the material quantity taken from the silo and the speed of the unloading motor and the real-time weighing data of the robot scale.

[0023] S6: Wait for the robot to actively request the feeding task, obtain the feeding task and the amount of feed taken from the silo, obtain the task process, determine the silo, sheep house, and feeding route based on the feeding path planning results, determine the current machine vehicle load weight based on the real-time weighing data of the robot scale, and control the robot to automatically perform the task.

[0024] The sheep feeding intelligent robot used in this embodiment is a dual-arm carrying mobile robot equipped with vision, radar, gravity sensor and other devices.

[0025] Furthermore, constructing a three-dimensional digital map of the sheep farm based on the three-dimensional point cloud data in S1 includes the following steps: S1.1: Use a laser radar scanning device to scan the sheep farm area with a laser beam and record the reflected signal to obtain point cloud data.

[0026] S1.2: Filter the point cloud data to obtain denoised point cloud data.

[0027] S1.3: Use the RANSAC algorithm to perform plane fitting on the point cloud data to identify the structures of the ground, walls, and railings, and remove point cloud data that is not on the ground or is not important.

[0028] S1.4: Based on the sheep farm geographic information, generate a three-dimensional data set by aggregating the processed point cloud data according to the spatial coordinate system, so that each point cloud data has the corresponding spatial coordinates and additional information of color and intensity; S1.5: Based on the 3D dataset, convert the point cloud data into a mesh, surface, or volume representation to construct a complete 3D digital map.

[0029] Furthermore, the present invention determines the life cycle stages of sheep, including the scalp stage, the young sheep stage and the adult sheep stage; the scalp stage is from birth to 6 months; the young sheep stage is from 6 months to 12 months; and the adult sheep stage is more than 12 months.

[0030] Determine the feeding amount required for each sheep at different weights in different life cycle stages and establish a sheep feed feeding amount database.

[0031] The present invention determines the feed silo takeout amount and feeding tasks for each sheep pen based on the number and weight data of sheep at different life cycle stages in each pen and the sheep feed feeding amount database, including the following steps: S2.1: Configure surveillance cameras to regularly collect surveillance images of each sheep pen; perform pre-processing on the surveillance images by denoising, cropping, and standardization.

[0032] S2.2: Use computer vision technology to identify the pre-processed monitoring images through a pre-trained convolutional neural network model, and determine the number of sheep and the corresponding life cycle stage based on the sheep's body shape and color feature data.

[0033] S2.3: Obtain the weight of sheep using ground pressure sensors and verify the number of sheep at different life cycle stages.

[0034] S2.4: Determine the feeding tasks and feed silo withdrawal amounts for each sheep pen based on the sheep feed intake database and the number of sheep at different life cycle stages in each sheep pen.

[0035] Furthermore, the locations of the feed silo and each sheep pen are determined based on the three-dimensional digital map. The feed silo is used as the starting point and the end point, and each sheep pen is used as a path point. The feeding path planning is carried out with the shortest path as the goal, including the following steps: S3.1: Based on the three-dimensional digital map, mark the location of the feed silos and sheep pens within the sheep farm area.

[0036] S3.2: Convert the coordinate information in the three-dimensional digital map into a two-dimensional coordinate system.

[0037] S3.3: Based on the two-dimensional coordinate system, taking into account obstacles and site limitations, the Dijkstra algorithm is used to calculate the shortest planar path from the silo to each sheep pen.

[0038] S3.4: Based on the location and number of sheep pens, use a greedy algorithm to optimize the order of visiting the sheep pens to ensure the shortest path.

[0039] This embodiment provides a feeding system based on an intelligent sheep feeding robot, such as Figure 2 As shown, the system includes: The data acquisition module is used to obtain three-dimensional point cloud data of the sheep farm area, collect monitoring images of each sheep pen in the sheep farm area, and collect weight data of the sheep in the sheep pen.

[0040] The Raspberry Pi controller is used to construct a 3D digital map of the sheep farm based on 3D point cloud data. It is also used to establish a database of sheep feed amounts at different life cycle stages. By identifying the number of sheep at different life cycle stages through monitoring images, the database determines the feed silo intake and feeding tasks for each sheep pen based on the number and weight data of sheep at different life cycle stages in each pen. It is also used to determine the location of the silo and each sheep pen based on the 3D digital map, using the silo as the starting and end point and each sheep pen as a route point to plan the feeding route with the shortest path as the goal.

[0041] The actuator is used to control the sheep feeding intelligent robot to perform feed hopper and feeding in each sheep pen along the planned feeding path based on the feed hopper feed volume and the feeding tasks of each sheep pen; the sheep feeding intelligent robot collects environmental information, performs obstacle avoidance, and returns to the initially set feeding path after avoiding obstacles until the planned feeding path is completed and feeding is completed.

[0042] The cloud server is used to store the amount of feed taken from the silo, the feeding tasks of each sheep pen, and the feeding path planning results, and conduct data analysis to determine the feeding strategy.

[0043] Mobile apps and farmers' PCs are used to obtain feed silo intake, feeding tasks for each sheep pen, feeding route planning results, and feeding strategies stored on the cloud server through wireless communication.

[0044] This system enables precise feeding of sheep, real-time monitoring of the farm environment, and optimal planning of feeding routes. It supports remote monitoring and operation, significantly reducing labor costs and enhancing the intelligence of animal husbandry management. The system uses a Raspberry Pi 5 development board as the main control unit, controlling multiple servo motors via the RS485 communication protocol, enabling flexible movement and precise feed delivery. First, the feeding route is planned. Point cloud mapping and laser 3D technology are used to map the farm's geographic information. 3D point cloud data is collected using technologies such as laser scanning and structured light scanning. Pre-processing operations such as filtering algorithms are then performed to generate a point cloud dataset and construct a 3D digital map. For precise feeding, a database of feed amounts for sheep at different lifecycle stages is established. The number of sheep in each pen is then used to inform the robot to perform a series of actions, including retrieving feed from the corresponding silo, calculating feed weight, and determining the location of the feeding pen. The system also integrates high-definition cameras and lidar to collect real-time environmental information, ensuring safe navigation in complex terrain. This data is then used to make intelligent path planning and feeding decisions. The system uses high-definition cameras for image recognition, laser radar for distance measurement, pressure sensors for calculating feed weight, ultrasonic sensors for obstacle detection, and a voice module to enhance human-computer interaction. The system is also connected to a cloud server, through which users can remotely monitor system status, view feeding data, and adjust feeding strategies via mobile apps and farmer PCs. This provides farmers with greater flexibility and convenience, facilitating precise management. Users can implement remote monitoring and data analysis to further optimize the farming management process.

[0045] The intelligent sheep-feeding robot proposed in this invention can automatically store, transport, and distribute feed according to a preset feeding plan, significantly improving the level of feeding automation. This not only reduces the time and labor required for manual feeding, but also makes the feeding process more punctual and accurate. Furthermore, the intelligent sheep-feeding robot can personalize feeding according to the growth stage and needs of the sheep, improving feed utilization and growth rate, thereby improving overall farming efficiency.

[0046] The application of the intelligent sheep-feeding robot proposed in this invention reduces farming costs. On the one hand, automated feeding reduces labor costs, enabling farmers to save on hiring labor. On the other hand, through precise feeding and personalized management, the intelligent sheep-feeding robot can reduce feed waste and improve feed conversion efficiency, thereby reducing feed costs.

[0047] The intelligent management capabilities of the sheep-feeding robot proposed in this invention make livestock farming management more refined and intelligent. The robot monitors sheep's health and environmental parameters in real time, providing scientific farming recommendations through big data analysis. This intelligent management approach not only improves the accuracy and efficiency of livestock farming but also helps farmers promptly identify and address problems that arise during the farming process, reducing risks. Furthermore, the robot can record and analyze livestock farming data, providing decision-making support for farmers and promoting the sustainable development of the livestock farming industry.

[0048] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A feeding method based on an intelligent sheep feeding robot, characterized in that: The following steps are involved: Determine the feeding tasks and feed silo intake for each sheep pen based on the sheep feed intake database and the number of sheep at different life cycle stages in each sheep pen; Determine feeding time and generate task flow based on feeding tasks of each sheep pen and feed silo intake; Scan the sheep farm area with LiDAR to generate a three-dimensional digital map; The location of the feed silo and each sheep pen is determined based on a three-dimensional digital map. The feed silo is used as the starting and ending point, and each sheep pen is used as a path point. The feeding route is planned with the shortest path as the goal. According to the dynamic algorithm, the current vehicle-loaded material weight is dynamically calculated based on the material quantity taken from the silo and the speed of the unloading motor and the real-time weighing data of the robot scale; Wait for the robot to actively request feeding tasks, obtain feeding tasks and feed silo feed quantity, obtain task process, determine the silo, sheep house, and feeding route based on the feeding path planning results, determine the current machine vehicle load weight based on the real-time weighing data of the robot scale, and control the robot to automatically perform the task.

2. The feeding method based on the sheep feeding intelligent robot according to claim 1, characterized in that: The method of scanning the sheep farm area by laser radar to generate a three-dimensional digital map includes the following steps: Use a laser radar scanning device to scan the sheep farm area with a laser beam and record the reflected signal to obtain point cloud data; Filter the point cloud data to obtain denoised point cloud data; Perform plane fitting on the point cloud data using the RANSAC algorithm to identify the structures of the ground, walls, and railings, and remove non-ground or non-important point cloud data; Based on the geographic information of the sheep farm, a three-dimensional data set is generated by aggregating the processed point cloud data according to the spatial coordinate system, so that each point cloud data has the corresponding spatial coordinates and additional information of color and intensity; Based on the 3D data set, the point cloud data is converted into mesh, surface or volume representation to construct a complete 3D digital map.

3. The feeding method based on the sheep feeding intelligent robot according to claim 1, characterized in that: The construction of the sheep feed feeding amount database comprises the following steps: Identify the life cycle stages of sheep, including the lamb stage, young lamb stage and adult lamb stage; the lamb stage is from birth to 6 months; the young lamb stage is from 6 months to 12 months; and the adult lamb stage is over 12 months; Determine the feeding amount required for each sheep at different weights in different life cycle stages and establish a sheep feed feeding amount database.

4. The feeding method based on the sheep feeding intelligent robot according to claim 3, characterized in that: The method of determining the feeding task and feed silo takeout amount of each sheep pen based on the sheep feed feeding amount database and the number of sheep in different life cycle stages of each sheep pen comprises the following steps: Configure surveillance cameras to regularly collect surveillance images of each sheep pen; perform pre-processing on surveillance images by denoising, cropping, and standardization; Using computer vision technology, a pre-trained convolutional neural network model is used to identify pre-processed surveillance images. Based on the sheep's body shape and color characteristics, the number of sheep and their corresponding life cycle stage are determined. The weight of sheep is obtained through ground pressure sensors to verify the number of sheep at different life cycle stages; Based on the sheep feed feeding amount database and the number of sheep in different life cycle stages in each sheep pen, the feeding task and feed silo intake of each sheep pen are determined.

5. The feeding method based on the sheep feeding intelligent robot according to claim 1, characterized in that: The method of determining the location of the feed silo and each sheep pen based on the three-dimensional digital map, taking the feed silo as the starting point and the end point, and each sheep pen as the path point, and planning the feeding path with the shortest path as the goal, includes the following steps: Based on the three-dimensional digital map, mark the location of the feed silos and sheep pens within the sheep farm area; The z-axis of the point cloud data is taken as interval values, compressed into a two-dimensional image, and the coordinate information in the three-dimensional digital map is converted into a two-dimensional coordinate system; Based on a two-dimensional coordinate system, taking into account obstacles and site restrictions, the coordinate positions of obstacles, feed troughs, and silos are marked, and the Dijkstra algorithm is used to calculate the shortest planar path from the silo to each sheep pen; Based on the location and number of sheep pens, a greedy algorithm is used to optimize the order of visiting the sheep pens to ensure the shortest path.

6. The feeding method based on the sheep feeding intelligent robot according to claim 1, characterized in that: When the sheep feeding intelligent robot is controlled along the planned feeding path to perform material retrieval from the feed bin and feeding in each sheep pen, the material retrieval amount and the feeding amount are determined by the pressure sensor data of the material storage area of ​​the sheep feeding intelligent robot.

7. A feeding system based on an intelligent sheep feeding robot, characterized in that: The system comprises: Data acquisition module, used to obtain three-dimensional point cloud data of the sheep farm area; The Raspberry Pi controller is used to construct a 3D digital map of the sheep farm based on 3D point cloud data. It is also used to establish a database of sheep feed requirements at different life cycle stages. Monitoring images are used to identify the number of sheep at different life cycle stages. Based on the sheep feed requirement database, the feed silo intake and feeding tasks for each pen are determined based on the number and weight data of sheep at different life cycle stages in each pen. The system is also used to determine the location of the silo and each pen based on the 3D digital map, using the silo as the starting and end point and each pen as a path point to plan a feeding route with the shortest path as the goal. A dynamic algorithm is also used to dynamically calculate the current vehicle-loaded material weight based on the feed silo intake and the real-time weighing data from the robot scale. The robot is used to obtain feeding tasks and feed silo feed quantity, obtain task processes, determine the silo, sheep house, and feeding route based on the feeding path planning results, determine the current machine vehicle load weight based on the real-time weighing data of the robot scale, and control the robot to automatically perform tasks.

8. The sheep feeding system based on the intelligent robot for sheep feeding according to claim 7, characterized in that: Also includes: The cloud server is used to store the amount of feed taken from the silo, the feeding tasks of each sheep pen, and the feeding path planning results, and conduct data analysis to determine the feeding strategy; Mobile apps and farmers' PCs are used to obtain feed silo intake, feeding tasks for each sheep pen, feeding route planning results, and feeding strategies stored on the cloud server through wireless communication.

Images