Unmanned harvesting system for radishes

Abstract

The invention provides an unmanned harvesting system for radishes, and relates to the technical field of agricultural artificial intelligence, and the system intelligently controls the movement of a radishes harvester through a radishes harvesting unmanned control subsystem, a radishes harvesting unmanned operation subsystem and a radishes agricultural parameter configuration subsystem. The shape of the vegetable field plot, the terrain and the planting agriculture are fused, and an optimal operation path is planned; on the basis of various parameter configurations, a computer vision algorithm is fused, and radish fruits and radish leaves are accurately identified. The radish unmanned harvesting system provided by the invention realizes more intelligent, fine and accurate operation of the radish unmanned operation harvester, so that manpower, material resources and time are saved, and the efficiency and quality of radish unmanned harvesting and the green and intelligent level of radish unmanned harvesting are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural artificial intelligence, and in particular to an unmanned radish harvesting system. Background Art

[0002] Radish is a root vegetable with extremely high edible and medicinal value. The harvesting process is the most time-consuming and labor-intensive, which usually includes digging, loosening the soil, conveying, cutting, and transportation.

[0003] The existing harvesters for intelligent harvesting of radishes focus on solving the problems of mechanized combined harvesting of radishes and the intelligence of some links, and connect the operation process of collecting, clipping, transporting and cutting radishes, which is structurally adapted to the radish production environment. However, the harvesting process often requires more than two people to collaborate, and has high professional requirements for harvesters and sorters, and cannot achieve unmanned intelligent harvesting of the entire process. In addition, in hot weather, using existing harvesters to harvest radishes requires extremely high physical strength and energy from the driver, and is extremely labor-intensive, material-intensive and time-consuming. Summary of the invention

[0004] The invention provides an unmanned radish harvesting system to solve the above problems existing in the prior art.

[0005] The present invention provides an unmanned radish harvesting system, comprising an unmanned radish harvesting control subsystem, an unmanned radish harvesting operation subsystem and a radish agronomic parameter configuration subsystem; The unmanned control subsystem for radish harvesting includes an unmanned mobile control module and an unmanned radish harvesting module; the unmanned mobile control module is mainly composed of a remote control module, a wireless communication system, a Beidou satellite positioning system and an electric crawler chassis; the unmanned radish harvesting module uses an angle sensor for height control, and infers the height of the working platform and the ridge through the angle; The unmanned radish harvesting subsystem includes a ridge navigation extraction module and a U-turn and line-changing module; the ridge navigation extraction module includes an RTK module and a visual module, the RTK module is used to use the radish agronomic parameter information, apply the composition information in the harvesting operation, and realize the position recognition of the radish; the visual module is used to perform ridge alignment through the recognition of the ridge ditch edge information, and realize real-time operation path planning; the U-turn and line-changing module is used to plan the path to the machine farming road, the field head, and the edge of the operation plot during the operation process; The radish agronomic parameter configuration subsystem includes a variety parameter configuration module, a cultivation parameter configuration module and an environmental parameter configuration module; the variety parameter configuration module is used to obtain variety parameter configuration information; the cultivation parameter configuration module is used to obtain cultivation parameter configuration information; the environmental parameter configuration module is used to obtain environmental parameter configuration information.

[0006] According to an unmanned radish harvesting system provided by the present invention, the unmanned mobile control module includes an unmanned speed control module and an unmanned steering control module; the unmanned speed control module adopts a servo to perform soft voltage regulation; the unmanned steering control module adopts proportional valve control.

[0007] According to the unmanned radish harvesting system provided by the present invention, the U-turn and line-changing module adopts the elite ant colony algorithm to perform path optimization.

[0008] According to an unmanned radish harvesting system provided by the present invention, the unmanned radish harvesting operation subsystem includes a pinching and plucking visual recognition module; the pinching and plucking visual recognition module adopts an instance segmentation algorithm to obtain radish fruits.

[0009] According to an unmanned radish harvesting system provided by the present invention, the segmentation network of the instance segmentation algorithm uses the Yolov8s-seg model as the backbone network; the backbone of the Yolov8s-seg model adopts MobileNetv4; the SPPF module in the Yolov8s-seg model is replaced by a large-core separation convolution attention module; and the segmentation network includes an inverted bottleneck search module.

[0010] According to the unmanned radish harvesting system provided by the present invention, the visual module uses the Hough operator and the Otsu method with adaptive threshold to binarize the collected images, and processes the loss function based on the Huber algorithm to identify the ridge center line and the boundaries on both sides of the ridge during the radish harvesting process.

[0011] According to the unmanned radish harvesting system provided by the present invention, the unmanned radish harvesting system further includes an edge server and a cloud server; The unmanned radish harvesting system adopts a cloud-edge-end intelligent service architecture; the unmanned radish harvesting control subsystem and the unmanned radish harvesting operation subsystem are implemented through terminals; and big data processing is processed on edge servers and / or cloud servers.

[0012] According to the unmanned radish harvesting system provided by the present invention, the cultivation parameter configuration information includes one or more of the following information: Overall operation area configuration; Planting ridge height; ridge spacing; Line spacing; Plant spacing; Cultivation time.

[0013] According to the unmanned radish harvesting system provided by the present invention, the variety parameter configuration information includes one or more of the following information: Radish varieties to grow: The size of the fruit contained; Time to mature; Water requirements at different stages; Fertilizer requirements at different stages; Suitable temperature; Suitable humidity.

[0014] According to the unmanned radish harvesting system provided by the present invention, the environmental parameter configuration information includes one or more of the following information: Air temperature; Air humidity; Soil temperature; Soil moisture; illumination; Soil pH.

[0015] The present invention provides an unmanned radish harvesting system. The system intelligently controls the movement of a radish harvester through an unmanned radish harvesting control subsystem, an unmanned radish harvesting operation subsystem and a radish agronomic parameter configuration subsystem, and integrates the shape of vegetable fields, topography and planting agronomy to plan the optimal operation path. Based on the configuration of multiple parameters, the system integrates computer vision algorithms to accurately identify radish fruits and radish leaves, thereby achieving more intelligent, delicate and precise operation of the unmanned harvester, thereby saving manpower, material resources and time, and improving the efficiency and quality of unmanned radish harvesting as well as the green and intelligent level of unmanned radish harvesting. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 The present invention provides a schematic diagram of the technical framework of an unmanned radish harvesting system.

[0018] Figure 2 The present invention provides a schematic diagram of intelligent transformation of a radish harvester in an unmanned radish harvesting system.

[0019] Figure 3 The present invention provides a schematic diagram of differential track tracking in an unmanned radish harvesting system.

[0020] Figure 4 The present invention provides a schematic diagram of a crawler chassis power module in an unmanned radish harvesting system.

[0021] Figure 5 The present invention provides a schematic diagram of radish path planning in an unmanned radish harvesting system.

[0022] Figure 6 The present invention provides a schematic diagram of the structure of an elite ant colony algorithm in an unmanned radish harvesting system.

[0023] Figure 7 The present invention provides an improved Yolov8s-seg network structure schematic diagram in an unmanned radish harvesting system.

[0024] Figure 8 It is a schematic diagram of the radish harvesting cloud-edge intelligent processing architecture in the unmanned radish harvesting system provided by the present invention. DETAILED DESCRIPTION

[0025] Radish is a root vegetable with extremely high edible and medicinal value. Currently, it is planted in a large area, and the harvesting stage is the most time-consuming and labor-intensive, which usually includes digging, loosening the soil, conveying, cutting, and transportation.

[0026] Existing intelligent radish harvesting machines, such as radish harvesters or radish combine harvesters, can perform deep loosening of the soil, leaf raking, tassel pulling, radish transportation, and fruit separation and cutting during the harvesting process, realizing multifunctional combined radish harvesting operations and having strong scene adaptability. However, during the radish harvesting process, the driver needs to focus on the radish clamping and pulling side for a long time, and at the same time needs to consider whether the radish is cut into fruits. There has been no unmanned transformation, and there is often a lack of systematic connection as a whole, and a lack of expansion of intelligent electronic control and automatic control.

[0027] Therefore, the existing technology focuses on solving the problems of mechanized combined harvesting of radishes and the intelligence of some links, and connects the operation process of radish collection, clamping, transmission, and cutting, which is structurally adapted to the radish production environment. However, the harvesting process requires more than two people to collaborate, and there are more professional requirements for harvesters and sorters. Even if some links are intelligently transformed and methods are provided to avoid fruit damage during radish harvesting, it still needs to rely on the quality of the clamping link of the collection device, and it is impossible to achieve a complete process of harvesting through intelligence, especially in hot weather, which requires extremely high physical strength and energy from the driver.

[0028] On this basis, the unmanned radish harvesting system proposed in the present invention integrates radish production agronomy, various types of radish harvesting agricultural machinery and data collected during the operation process. It can be based on the radish harvester for unmanned transformation, and can intelligently define the forward and backward movement, direction, speed, and lifting of the harvesting arm of the harvester. It integrates the shape of the vegetable field, terrain and planting agronomy, and plans the optimal operation path to save fuel and protect the microenvironment of the vegetable field, integrating the concept of green production into the unmanned driving model.

[0029] In addition, the unmanned radish harvesting system proposed in the present invention integrates computer vision algorithms in multiple links to accurately identify radish fruits and radish leaves, and realizes more intelligent and delicate operation of the harvester by the unmanned operation model, providing support for the development of new quality productivity in the vegetable industry.

[0030] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are 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.

[0031] The unmanned radish harvesting system proposed in the present invention has been unmanned for radish harvesting equipment, and an overall transformation method covering power module, direction control module, harvesting operation module, path planning module, precise ridge alignment module, and fruit recognition module has been proposed to adapt to the unmanned harvesting scene of standardized radish planting. All models of unmanned operation transformation adopt cloud-edge processing architecture, and the computing power overhead is minimized through model scenario adaptation scheduling. In the path planning, precise ridge alignment, and fruit recognition modules of the harvesting process, the need to save fuel and electricity consumption is considered from the aspects of the shortest path and the best computing power. In the process of realizing unmanned operation, carbon emissions can also be reduced, and the realization of green, efficient and intelligent production of vegetables can be promoted.

[0032] Combine the following Figures 1 to 8 The present invention describes the unmanned radish harvesting system.

[0033] Figure 1 The following is a technical framework diagram of an unmanned radish harvesting system provided by the present invention. Figure 1 As shown, the system includes an unmanned control subsystem for radish harvesting, an unmanned operation subsystem for radish harvesting, and a radish agronomic parameter configuration subsystem; The unmanned control subsystem for radish harvesting includes an unmanned mobile control module and an unmanned radish harvesting module; the unmanned mobile control module is mainly composed of a remote control module, a wireless communication system, a Beidou satellite positioning system and an electric crawler chassis; the unmanned radish harvesting module uses an angle sensor for height control, and infers the height of the working platform and the ridge through the angle; The unmanned radish harvesting subsystem includes a ridge navigation extraction module and a U-turn and line-changing module; the ridge navigation extraction module includes an RTK module and a visual module, the RTK module is used to use the radish agronomic parameter information, apply the composition information in the harvesting operation, and realize the position recognition of the radish; the visual module is used to perform ridge alignment through the recognition of the ridge ditch edge information, and realize real-time operation path planning; the U-turn and line-changing module is used to plan the path to the machine farming road, the field head, and the edge of the operation plot during the operation process; The radish agronomic parameter configuration subsystem includes a variety parameter configuration module, a cultivation parameter configuration module and an environmental parameter configuration module; the variety parameter configuration module is used to obtain variety parameter configuration information; the cultivation parameter configuration module is used to obtain cultivation parameter configuration information; the environmental parameter configuration module is used to obtain environmental parameter configuration information.

[0034] Specifically, the embodiment of the present invention takes harvesting white radish as an example to illustrate the unmanned radish harvesting system of the present invention. Figure 2 The figure is a schematic diagram of intelligent transformation of a radish harvester in an unmanned radish harvesting system provided by the present invention. Figure 2 As shown. The radish harvester in the radish unmanned harvesting system provided by the embodiment of the present invention is based on a clamping radish harvester, and the basic structure includes a double-row radish tassel-supporting device, a fruit cutting device, a crawler power device, a radish fruit transmission device, a transverse conveyor belt, a driving control room, a radish tassel-supporting electric control switch, a forward and backward power joystick, a harvesting operation joystick, and the radish machine direction and harvesting arm are adjusted up and down. A navigation antenna, a millimeter-wave radar, a laser radar, a binocular camera, and a touch screen are configured on the radish harvesting platform, wherein the navigation antenna is spaced 1 meter apart at equal heights, and the touch screen provides an intelligent control entrance for the radish harvester, and integrates the multimodal data of the millimeter-wave radar, the laser radar, and the binocular camera for mapping and obstacle avoidance. The joysticks are all connected to the steering gear electronic control module, including controlling the direction and speed of the operating platform and the lifting and lowering of the radish harvesting arm, and unmanned transformation and improvement are carried out on this basis.

[0035] The unmanned transformation of the radish harvesting equipment (i.e., radish harvester) provided by the embodiment of the present invention integrates the electronic control technology, Beidou navigation technology, computer vision recognition technology, and control feedback technology and applies them to the power control, operation tool control, operation path planning, precise row alignment, and fruit cutting of the radish harvesting equipment, and performs systematic adaptation.

[0036] Optionally, the unmanned mobile control module includes an unmanned speed control module and an unmanned steering control module; the unmanned speed control module adopts a servo to perform soft voltage regulation; the unmanned steering control module adopts proportional valve control.

[0037] Specifically, the unmanned speed control module uses a servo to softly regulate the voltage, making the radish harvester move more stably and reducing vibrations during speed changes.

[0038] The unmanned steering control module adopts proportional valve control, which improves the accuracy of steering control and facilitates the radish harvester to change direction flexibly.

[0039] The unmanned steering control module is mainly composed of a remote control module, a wireless communication system, a Beidou satellite positioning system and an electric track chassis.

[0040] The radish harvester uses millimeter-wave radar, lidar, and visual sensor fusion identification to monitor obstacles in real time during the moving and reversing processes.

[0041] In the embodiment of the present invention, in order to realize automatic control of the path of the tracked chassis of the radish harvester during the radish harvesting process, it is necessary to define the posture of the radish harvester during its movement. Figure 3 FIG. 1 is a schematic diagram of differential track tracking in an unmanned radish harvesting system provided by the present invention. Figure 3 Take the center of the radish harvester track as the base point of the motion coordinates, and the center point of the track chassis is O B (x, y), the velocity is , the speed of the left track is , the speed of the right track is , O B The angle between the direction of the point velocity and the positive x-axis of the coordinate system is the heading angle of the tracked chassis, and the heading angle is positive when it is counterclockwise from the positive direction of the x-axis coordinate. Figure 4 The schematic diagram of the crawler chassis power module in the unmanned radish harvesting system provided by the present invention is as follows: Figure 4 shown.

[0042] According to the principle that speed is equal to displacement derivative, the speed is decomposed in the x-direction and y-direction. According to the vector parallelogram decomposition law, the kinematic equation of the crawler chassis can be obtained as follows: (1) (2) Since the track chassis can be regarded as a rigid structure, the movement of the center points of the left and right tracks at any instant in the motion process can be regarded as circular motion around the same center of a circle, so the angular velocity of the center points of the left and right tracks is equal to the angular velocity of the center point of the track , we can get the angular velocity of the center points of the left and right tracks for: (3) in, Indicates the distance between the center points of the left and right tracks. Figure 3 .

[0043] Combining equations (1) to (3), the kinematic model of the tracked chassis is as follows: (4) For unmanned radish harvesting, the planned path is usually composed of a series of points. The path points contain information such as the posture of the harvesting equipment and the driving speed. If the machine follows a point, it can be regarded as following the curved path. The radish harvesting speed is usually below 3km / h. For low-speed motion, a pure tracking algorithm is suitable. The turning radius is solved by set sum. Establish the coordinate system X B -O B -Y B (Continued see Figure 3 ), the center point O of the track model B The coordinates are (x, y), the target point P B The coordinates are (x b ,y b ), in order to allow the crawler chassis to move from O B Point movement to P B point, set the differential speed of the left and right crawler tracks of the crawler chassis to be fixed, so that O B Point moves to P along an arc trajectory B Point. Set point O C is the center of the turning curve.

[0044] Define the coordinate origin O B With the target point P B The straight line where B The angle between the axes is , LE is the forward distance of the motion preview, R is the turning radius, and β is defined as the arc angle of the track chassis from the initial position to the target point. According to the geometric relationship, we have: (5) We can get: (6) (7) The target point P B Go to XB Projection on the axis, according to the geometric relationship, we can get: (8) (9) Combining equations (6) to (9), the turning radius of the crawler chassis is obtained as: (10) Assuming that the crawler chassis is in pure friction with the ground when in motion (i.e., no slipping is considered), the speed of the left and right crawlers is , Chassis center speed There is the following inherent relationship: (11) (12) Substituting (10) into (11) and (12), the given value of the differential track speed can be finally solved as follows: (13) Define x b is the lateral error of path tracking, and α is the heading deviation of path tracking. b The larger the absolute value, the greater the lateral distance that the harvester deviates from the target path; when the absolute value of the angle between the target point and the x-axis of the crawler chassis coordinate system is larger, it means that the heading error of the harvester deviates from the target path is larger.

[0045] The unmanned control subsystem for radish harvesting provided by the embodiment of the present invention improves the forward and backward travel and left and right steering devices of the radish harvester, adopts a servo to softly control the voltage to solve the problems of abrupt speed switching of the original mechanical travel and unstable turning due to locking of one-side track, and simultaneously introduces electronic control to provide on-site remote control and remote algorithm control ports, thereby realizing effective control of the radish harvester.

[0046] The unmanned operation subsystem for radish harvesting includes a ridge navigation extraction module and a U-turn and line-changing module, wherein the ridge navigation extraction module includes an RTK module and a vision module.

[0047] The RTK module mainly uses the basic information of the radish cultivation process and applies the composition information in the harvesting operation to realize the position identification of the radish.

[0048] The visual module is mainly used for ridge alignment. By identifying the ridge edge information, it can achieve real-time operation path planning with greater accuracy.

[0049] Optionally, the visual module uses the Hough operator and the Otsu method with an adaptive threshold to binarize the collected image, and processes the loss function based on the Huber algorithm to identify the ridge center line and the ridge side boundaries during the radish harvesting process.

[0050] Specifically, when radish ridge alignment is performed based on the vision module, the Hough operator and OTSU with adaptive threshold are used for binarization, and the loss function is processed based on the Huber algorithm to identify the ridge centerline and ridge side boundaries during radish harvesting, thereby achieving accurate ridge alignment during radish harvesting.

[0051] Optionally, the U-turn module uses an elite ant colony algorithm to perform path optimization.

[0052] Specifically, the U-turn module is mainly used to plan the path to the mechanized farming road, the edge of the field, and the edge of the working plot during the operation process. The embodiment of the present invention adopts the elite ant colony algorithm to optimize the path.

[0053] Figure 5 FIG. 1 is a schematic diagram of radish path planning in an unmanned radish harvesting system provided by the present invention. Figure 5 shown.

[0054] Figure 6 Schematic diagram of the structure of an elite ant colony algorithm in an unmanned radish harvesting system provided by the present invention. Figure 6 shown.

[0055] Based on any of the above embodiments, the radish harvesting process uses various operators, algorithms and computer vision models in the process of operation path planning, navigation, obstacle identification, ridge alignment, clipping, cutting, and radish quality detection. Since the model for agricultural production scene services usually needs to have high precision and portability, the embodiment of the present invention is designed in two aspects: path planning and target detection.

[0056] Taking the fruit position detection in the radish pulling process as an example, the elite ant colony algorithm is used for path planning to optimize the path; in terms of target detection, the segmentation algorithm is used to accurately acquire the target so that the machine can perform rapid feedback and operation.

[0057] Optionally, the unmanned radish harvesting subsystem includes a pinching and plucking visual recognition module; the pinching and plucking visual recognition module uses an instance segmentation algorithm to obtain radish fruits.

[0058] The segmentation network of the instance segmentation algorithm uses the Yolov8s-seg model as the backbone network; the backbone of the Yolov8s-seg model uses MobileNetv4; the SPPF module in the Yolov8s-seg model is replaced by a large-core separation convolution attention module; and the segmentation network includes an inverted bottleneck search module.

[0059] Specifically, the radish harvesting unmanned operation subsystem also includes a clipping and pulling visual recognition module and a radish fruit and tassel visual recognition module. The clipping and pulling visual recognition module is used to identify the clipping and pulling position of the radish so as to accurately clip and pull the radish, and the radish fruit and tassel visual recognition module is used to identify the position of the radish fruit and radish tassel so as to accurately cut the radish tassel.

[0060] In order to take into account recognition accuracy, low latency and small parameters (i.e., lower network parameter amount), the embodiment of the present invention uses the instance segmentation algorithm Yolov8s-seg that supports images and videos as the backbone network in the pinching visual recognition module and the radish fruit and tassel visual recognition module, and makes improvements on this basis. Figure 7 FIG. 1 is a schematic diagram of an improved Yolov8s-seg network structure in an unmanned radish harvesting system provided by the present invention, such as Figure 7 shown.

[0061] In order to increase the portability of the network, the backbone adopts MobileNetv4 and introduces a flexible inverted bottleneck search module. In addition, it integrates the inverted bottleneck search module, ConvNext, feedforward network and an additional deep separable variant, improves the model search efficiency through the optimized NAS formula, introduces novel distillation technology to improve model accuracy, and uses the large kernel separation convolution attention module (LSKA) to replace the SPPF module in the Yolov8s-seg model to improve the small target detection performance during the operation, thereby reducing the computational complexity and video memory usage.

[0062] The radish harvesting unmanned operation subsystem provided by the embodiment of the present invention can perform unmanned harvesting operation path planning. For the radish harvesting plot scene, the radish harvester's own mechanism and the position of the harvesting arm are integrated, and an improved ant colony algorithm is used for iterative calculation. Before each operation, the operation starting point, end point, and operation minimum energy consumption path planning are considered, thereby realizing the minimum energy consumption operation mode analysis for a fixed plot; it can also perform radish clipping and pulling image recognition, and adopt an improved Yolov8s-seg model to recognize the position of radish fruits in real time, use MobileNetv4 to replace the backbone network, and adopt LSKA to replace the SPPF module, so as to realize a lightweight model training process with few parameters, improve the performance of small target detection during the operation process, and reduce the computational complexity and video memory occupancy.

[0063] In the unmanned radish harvesting system provided by the present invention, the radish agronomic parameter configuration subsystem is a key link in the integration of agricultural machinery and agronomy. By entering agronomic parameters, driving requirements are provided for the harvester, including variety parameter configuration module, cultivation parameter configuration module and environment parameter configuration module.

[0064] Optionally, the variety parameter configuration information includes one or more of the following information: Radish varieties to grow: The size of the fruit contained; Time to mature; Water requirements at different stages; Fertilizer requirements at different stages; Suitable temperature; Suitable humidity.

[0065] Specifically, the variety parameter configuration module mainly inputs the planting radish variety, which contains information such as fruit size, maturity time, water requirement at different stages, fertilizer requirement, suitable temperature, humidity, etc.

[0066] Optionally, the cultivation parameter configuration information includes one or more of the following information: Overall operation area configuration; Planting ridge height; ridge spacing; Line spacing; Plant spacing; Cultivation time.

[0067] Specifically, cultivation parameters are the focus of the configuration process, including information such as the overall operating area configuration, planting ridge height, ridge spacing, row spacing, plant spacing, and cultivation time, which can be used as scene parameter input for unmanned agricultural machinery operations. In terms of agricultural machinery parameters, the wheelbase, agricultural machinery position, starting point, and operating mode are configured.

[0068] Optionally, the environmental parameter configuration information includes one or more of the following information: Air temperature; Air humidity; Soil temperature; Soil moisture; illumination; Soil pH.

[0069] Specifically, the environmental parameter configuration module is mainly aimed at the basic information of radish varieties, and configures suitable air temperature, air humidity, soil temperature, soil humidity, light, and soil pH.

[0070] Based on any of the above embodiments, the embodiment of the present invention uses computer vision to identify the ridge edge of the operation. The recognition model architecture refers to the improved Yolov8s-seg (continue to refer to Figure 7 ) method, the data input and annotation of the model are the edge lines of the radish ridges. In this scenario, the harvesting arm is on the right, so it is necessary to identify the left ridge line after the ridge, that is, to ensure the accuracy of the unmanned driving through real-time visual correction.

[0071] Optionally, the unmanned radish harvesting system further includes an edge server and a cloud server; The unmanned radish harvesting system adopts a cloud-edge-end intelligent service architecture; the unmanned radish harvesting control subsystem and the unmanned radish harvesting operation subsystem are implemented through terminals; and big data processing is processed on edge servers and / or cloud servers.

[0072] Specifically, Figure 8 : is a schematic diagram of a radish harvesting cloud-edge intelligent processing architecture in an unmanned radish harvesting system provided by the present invention, such as Figure 8 As shown. The unmanned radish harvesting system in the embodiment of the present invention adopts a "cloud, edge, and end" collaborative architecture for model training, data transmission, and model recognition. The model training process usually requires services with high computing power and high processing performance. Therefore, in the "cloud", it is implemented together by the control node of the cloud part Kubernetes and the node run by the edge part KubeEdge. The Kubernetes control node uses the original data model of the cloud part and keeps the original control and data flow unchanged, that is, the node run by KubeEdge appears as an ordinary node on Kubernetes. Kubernetes can manage the nodes run by KubeEdge like ordinary nodes. Based on the Kubernetes control node, KubeEdge realizes the sinking of Kubernetes cloud computing orchestration containerized applications through the CloudCore of the cloud part and the EdgeCore of the edge part. Here, after periodically collecting on-site training data, it is labeled and trained according to demand, a trained model is generated, and scheduling is performed.

[0073] The "edge" and "end" collaboration refers to using KubeEdge as a management program running on edge nodes, responsible for managing the resources, operation terminal operation status and faults of the unmanned operation process load on the edge nodes. KubeEdge provides the required computing resources for the EdgeXFoundry service, and is responsible for managing the entire life cycle of the EdgeXFoundry end service. EdgeXFoundry can collect, filter, store and mine data from a variety of IoT terminal devices through the managed harvesting operation microservice module, and can also control the terminal devices by issuing instructions to a variety of IoT terminal devices through the managed microservices. Through edge computing, model computing power is deployed for production scenarios, and scenario-based fine-tuning is performed for different tasks; embedded terminals are connected through high-bandwidth, low-latency networks, including video monitoring, obstacle avoidance terminals, navigation terminals, etc. at various locations in the radish harvester.

[0074] The cloud-edge intelligent service framework in the unmanned radish harvesting system provided by the embodiment of the present invention, in the process of executing unmanned operation tasks, the sensors, controllers, video collectors, mechanical posture controllers, etc. involved all follow the cloud-edge intelligent service architecture, that is, data is collected through the terminal at the operation site, small computing power data analysis and processing is completed locally, and the data is compressed before transmission. Finally, the high-energy consumption and large-parameter model training is set to be carried out in the cloud service, and the computing power overhead is reduced to the greatest extent through model scenario adaptation scheduling.

[0075] The present invention provides an unmanned radish harvesting system, which intelligently defines and controls the forward and backward movement, direction, speed, and lifting and lowering of a harvesting arm of a radish harvester through an unmanned control subsystem for radish harvesting, an unmanned operation subsystem for radish harvesting, and a radish agronomic parameter configuration subsystem, and integrates the shape of vegetable fields, topography, and planting agronomy to plan the optimal operation path; integrates computer vision algorithms in multiple links to accurately identify radish fruits and radish leaves, thereby achieving more intelligent, delicate, and precise operation of the unmanned harvester; all models transformed for unmanned operations adopt a cloud-edge processing architecture, and reduce computing power consumption through model scenario adaptation scheduling; thereby saving manpower, material resources, and time, and improving the efficiency and quality of unmanned radish harvesting, as well as the green and intelligent level of unmanned radish harvesting.

[0076] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0077] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0078] "Determine B based on A" in the embodiments of the present application means that the factor A should be considered when determining B. It is not limited to "B can be determined based on A alone", but should also include: "Determine B based on A and C", "Determine B based on A, C and E", "Determine C based on A, and further determine B based on C", etc. In addition, it can also include taking A as a condition for determining B, for example, "When A meets the first condition, use the first method to determine B"; for another example, "When A meets the second condition, determine B", etc.; for another example, "When A meets the third condition, determine B based on the first parameter", etc. Of course, it can also be a condition that takes A as a factor for determining B, for example, "When A meets the first condition, use the first method to determine C, and further determine B based on C", etc.

[0079] In the present invention, the term "plurality" refers to two or more than two, and other quantifiers are similar thereto.

[0080] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An unmanned radish harvesting system, characterized in that: It includes the unmanned control subsystem for radish harvesting, the unmanned operation subsystem for radish harvesting and the agronomic parameter configuration subsystem for radish; The unmanned control subsystem for radish harvesting includes an unmanned mobile control module and an unmanned radish harvesting module; the unmanned mobile control module is mainly composed of a remote control module, a wireless communication system, a Beidou satellite positioning system and an electric crawler chassis; the unmanned radish harvesting module uses an angle sensor for height control, and infers the height of the working platform and the ridge through the angle; The unmanned radish harvesting subsystem includes a ridge navigation extraction module and a U-turn and line-changing module; the ridge navigation extraction module includes an RTK module and a visual module, the RTK module is used to use the radish agronomic parameter information, apply the composition information in the harvesting operation, and realize the position recognition of the radish; the visual module is used to perform ridge alignment through the recognition of the ridge ditch edge information, and realize real-time operation path planning; the U-turn and line-changing module is used to plan the path to the machine farming road, the field head, and the edge of the operation plot during the operation process; The radish agronomic parameter configuration subsystem includes a variety parameter configuration module, a cultivation parameter configuration module and an environmental parameter configuration module; the variety parameter configuration module is used to obtain variety parameter configuration information; the cultivation parameter configuration module is used to obtain cultivation parameter configuration information; the environmental parameter configuration module is used to obtain environmental parameter configuration information.

2. The unmanned radish harvesting system according to claim 1, characterized in that: The unmanned mobile control module includes an unmanned speed control module and an unmanned steering control module; the unmanned speed control module adopts a servo to perform soft voltage regulation; the unmanned steering control module adopts proportional valve control.

3. The unmanned radish harvesting system according to claim 1, characterized in that: The U-turn module uses the elite ant colony algorithm to perform path optimization.

4. The unmanned radish harvesting system according to claim 1, characterized in that: The unmanned radish harvesting operation subsystem includes a pinching and plucking visual recognition module; the pinching and plucking visual recognition module uses an instance segmentation algorithm to obtain radish fruits.

5. The unmanned radish harvesting system according to claim 4, characterized in that: The segmentation network of the instance segmentation algorithm uses the Yolov8s-seg model as the backbone network; the backbone of the Yolov8s-seg model uses MobileNetv4; the SPPF module in the Yolov8s-seg model is replaced by a large-core separation convolution attention module; and the segmentation network includes an inverted bottleneck search module.

6. The unmanned radish harvesting system according to claim 1, characterized in that: The visual module uses the Hough operator and the Otsu method with an adaptive threshold to binarize the collected images, and processes the loss function based on the Huber algorithm to identify the ridge center line and the boundaries on both sides of the ridge during the radish harvesting process.

7. The unmanned radish harvesting system according to claim 1, characterized in that: The unmanned radish harvesting system also includes an edge server and a cloud server; The unmanned radish harvesting system adopts a cloud-edge-end intelligent service architecture; the unmanned radish harvesting control subsystem and the unmanned radish harvesting operation subsystem are implemented through terminals; and big data processing is processed on edge servers and / or cloud servers.

8. The unmanned radish harvesting system according to claim 1, characterized in that: The cultivation parameter configuration information includes one or more of the following information: Overall operation area configuration; Planting ridge height; ridge spacing; Line spacing; Plant spacing; Cultivation time.

9. The unmanned radish harvesting system according to claim 1, characterized in that: The variety parameter configuration information includes one or more of the following information: Radish varieties to grow: The size of the fruit contained; Time to mature; Water requirements at different stages; Fertilizer requirements at different stages; Suitable temperature; Suitable humidity.

10. The unmanned radish harvesting system according to claim 1, characterized in that: The environmental parameter configuration information includes one or more of the following information: Air temperature; Air humidity; Soil temperature; Soil moisture; illumination; Soil pH.

Images