Unmanned harvesting system for radish

Abstract

The present application provides a kind of radish unmanned harvesting system, it is related to agricultural artificial intelligence technical field, the system is through radish harvesting unmanned control subsystem, radish harvesting unmanned operation subsystem and radish agronomic parameter configuration subsystem, the movement of radish harvesting machine is intelligently controlled, and the optimal operation path planning is carried out by fusing vegetable field block shape, terrain and planting agronomy;On the basis of a variety of parameter configurations, accurate identification of radish fruit and radish flag is carried out by fusing computer vision algorithm.The radish unmanned harvesting system provided by the present application realizes more intelligent, delicate and accurate operation of radish unmanned operation harvester, so as to save manpower, material resources and time, and improve the radish unmanned harvesting efficiency, quality and the greenness, intelligent level of radish unmanned harvesting.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of agricultural artificial intelligence technology, and particularly relates to a radish unmanned harvesting system. BACKGROUND

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

[0003] The existing harvesting machine for intelligent harvesting of radish mainly solves the problems of mechanized and combined harvesting of radish, intelligentization of part of the process, and operation process of radish collection, clamping, transmission and cutting link, and adapts to the radish production environment in structure, but the harvesting process often requires more than two people to cooperate, and the harvesting personnel and sorting personnel have high professional requirements, and the complete process cannot be realized unmanned intelligent harvesting. In addition, in hot weather, the use of the existing harvesting machine for radish harvesting requires high physical and mental strength of the driver, and is extremely time-consuming in terms of manpower, material resources and time. SUMMARY

[0004] The present application provides a radish unmanned harvesting system to solve the above problems in the prior art.

[0005] The present application provides a radish unmanned harvesting system, which comprises a radish harvesting unmanned control subsystem, a radish harvesting unmanned operation subsystem and a radish agronomic parameter configuration subsystem.

[0006] The radish harvesting unmanned control subsystem comprises an unmanned movement control module and an unmanned radish harvesting module; the unmanned movement control module is mainly composed of a remote control module, a wireless communication system, a Beidou satellite positioning system and an electric caterpillar chassis; the unmanned radish harvesting module uses an included angle sensor for height control, and the height of the operation platform and the ridge is calculated by the included angle.

[0007] The radish harvesting unmanned operation subsystem comprises a ridge navigation extraction module and a U-turn and row changing module; the ridge navigation extraction module comprises an RTK module and a vision module, the RTK module is used to apply the layout information in the harvesting operation by using the radish agronomic parameter information, and the position of the radish is recognized; the vision module is used to navigate the ridge by recognizing the ridge ditch edge information, and realize real-time operation path planning; the U-turn and row changing module is used to plan the path of the machine tillage way, the land head and the edge of the operation plot during the operation process.

[0008] The radish agricultural parameter configuration subsystem includes a variety parameter configuration module, a cultivation parameter configuration module and an environment parameter configuration module; the variety parameter configuration module is used for obtaining variety parameter configuration information; the cultivation parameter configuration module is used for obtaining cultivation parameter configuration information; and the environment parameter configuration module is used for obtaining environment parameter configuration information.

[0009] According to the radish unmanned harvesting system provided by the application, the unmanned speed control module adopts the form of a rudder engine for voltage soft regulation and control; and the unmanned steering control module adopts a proportional valve control.

[0010] According to the radish unmanned harvesting system provided by the application, the U-turn and row changing module adopts an elite ant colony algorithm for path optimization.

[0011] According to the radish unmanned harvesting system provided by the application, the radish harvesting unmanned operation subsystem includes a clamping and pulling visual recognition module; the clamping and pulling visual recognition module adopts an instance segmentation algorithm to obtain radish fruits.

[0012] According to the radish unmanned harvesting system provided by the application, the segmentation network of the instance segmentation algorithm takes a Yolov8s-seg model as a backbone network; the backbone of the Yolov8s-seg model adopts a MobileNetv4; an SPPF module in the Yolov8s-seg model is replaced by a large-core separation convolution attention module; and the segmentation network includes an inverted bottle-neck search module.

[0013] According to the radish unmanned harvesting system provided by the application, the visual module respectively utilizes a Hough operator and an adaptive threshold Otsu method to binarize the collected images, and processes a loss function based on a Huber algorithm to identify a ridge center line and ridge side boundaries in a radish harvesting process.

[0014] According to the radish unmanned harvesting system provided by the application, the radish unmanned harvesting system further includes an edge server and a cloud server.

[0015] The radish unmanned harvesting system adopts a cloud edge intelligent service architecture; the radish harvesting unmanned control subsystem and the radish harvesting unmanned operation subsystem are realized through a terminal; and big data processing is processed in an edge server and / or a cloud server.

[0016] According to the radish unmanned harvesting system provided by the application, the cultivation parameter configuration information includes one or more of the following information:

[0017] An overall operation area configuration.

[0018] Planting ridges;

[0019] Row spacing;

[0020] line spacing;

[0021] Plant spacing;

[0022] Cultivation time.

[0023] According to the present invention, an unmanned radish harvesting system is provided, wherein the variety parameter configuration information includes one or more of the following:

[0024] Radish varieties to be planted:

[0025] The size of the fruit inside;

[0026] Maturity time;

[0027] Water requirements at different stages;

[0028] Fertilizer requirements at different stages;

[0029] Suitable temperature;

[0030] Suitable humidity.

[0031] According to the unmanned radish harvesting system provided by the present invention, the environmental parameter configuration information includes one or more of the following:

[0032] Air temperature;

[0033] Air humidity;

[0034] Soil temperature;

[0035] Soil moisture;

[0036] illumination;

[0037] Soil pH.

[0038] This invention provides an unmanned radish harvesting system. Through an unmanned radish harvesting control subsystem, an unmanned radish harvesting operation subsystem, and a radish agronomic parameter configuration subsystem, the system intelligently controls the movement of the radish harvester. It integrates the shape of the vegetable field plots, terrain, and planting agronomy to plan the optimal operation path. Based on multiple parameter configurations, it incorporates computer vision algorithms to accurately identify radish fruits and radish tops, achieving more intelligent, delicate, and precise operation of the unmanned harvester. This saves manpower, resources, and time, and improves the efficiency, quality, and green and intelligent level of unmanned radish harvesting. Attached Figure Description

[0039] In order to make the technical solutions in the present application or the prior art clearer, the accompanying drawings needed in the embodiments or the prior art description will be briefly introduced below. Obviously, the accompanying drawings in the following description are some embodiments of the present application, and other accompanying drawings can also be obtained by those skilled in the art without creative effort on the basis of these accompanying drawings.

[0040] Figure 1 It is a technical framework schematic diagram of a radish unmanned harvesting system provided by the present application.

[0041] Figure 2 It is an intelligent transformation schematic diagram of a radish harvesting machine in a radish unmanned harvesting system provided by the present application.

[0042] Figure 3 It is a differential track tracking schematic diagram in a radish unmanned harvesting system provided by the present application.

[0043] Figure 4 It is a tracked chassis power module schematic diagram in a radish unmanned harvesting system provided by the present application.

[0044] Figure 5 It is a radish path planning schematic diagram in a radish unmanned harvesting system provided by the present application.

[0045] Figure 6 It is an elite ant colony algorithm structure schematic diagram in a radish unmanned harvesting system provided by the present application.

[0046] Figure 7 It is an improved Yolov8s-seg network structure schematic diagram in a radish unmanned harvesting system provided by the present application.

[0047] Figure 8 It is a radish harvesting cloud edge intelligent processing architecture schematic diagram in a radish unmanned harvesting system provided by the present application. DETAILED DESCRIPTION

[0048] Radish is a root vegetable with very high edible and medicinal value, and at present, the planting area is wide, wherein the harvesting link is the most time-consuming and labor-intensive, which usually includes the processes of digging, loosening soil, conveying, cutting, transporting and the like.

[0049] The existing intelligent harvesting machine for radish, such as a radish harvester or a radish combine harvester, can perform soil subsoiling, leaf raking, leaf plucking, radish conveying and fruit separation and cutting during the harvesting process, realize multifunctional combined radish harvesting operation, and have strong scene adaptability, but the driver needs to pay attention to the radish clamping and plucking side for a long time, and needs to consider whether the radish is cut, and the unmanned modification is not performed, and the system connection is generally lacking, and the intelligent electric control and automatic control extension are lacking.

[0050] Therefore, the existing technology focuses on solving the problems of radish mechanized combined harvesting and intelligentization of part links, and realizes the operation process of radish folding, clamping, conveying and cutting, and adapts the radish production environment from the structure, but more than two people are needed to cooperate during the harvesting process, and the harvesting and sorting personnel have relatively professional requirements. Even if the local link is intelligently modified to provide a method for avoiding fruit damage during radish harvesting, the quality of the folding device clamping link cannot be ensured, and the complete process cannot be harvested through intelligence, especially in hot weather, which requires high physical and mental strength of the driver.

[0051] On this basis, the radish unmanned harvesting system provided by the present application integrates radish production agronomy, various radish harvesting agricultural machines and collected data during the operation process, can be unmanned modified based on the radish harvester, intelligently defines the forward and backward movement, direction, speed and lifting of the harvesting arm of the harvester, integrates the shape, terrain and planting agronomy of the vegetable field, plans the optimal operation path, saves fuel consumption to protect the vegetable field environment, and integrates the concept of green production into the unmanned driving model.

[0052] In addition, the radish unmanned harvesting system provided by the present application integrates computer vision algorithms in multiple links to accurately identify radish fruits and radish leaves, realizes more intelligent and delicate operation of the unmanned operation model on the harvester, and provides support for the development of new quality productivity of the vegetable industry.

[0053] In order to make the purpose, technical scheme and advantages of the present application clearer, the technical scheme in the present application will be described clearly and completely in combination with the drawings in the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor belong to the scope of protection of the present application.

[0054] The present application provides a carrot unmanned harvesting system, which is characterized by the following technical solutions.

[0055] The technical framework of the carrot unmanned harvesting system is shown in the following figure. Figures 1 to 8 The carrot unmanned harvesting system is described in the following.

[0056] Figure 1 The technical framework of the carrot unmanned harvesting system is shown in the following figure. Figure 1 The system includes a carrot harvesting unmanned control subsystem, a carrot harvesting unmanned operation subsystem, and a carrot agronomic parameter configuration subsystem.

[0057] The carrot harvesting unmanned control subsystem includes an unmanned mobile control module and an unmanned carrot 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 caterpillar chassis.

[0058] The carrot harvesting unmanned operation subsystem includes a row navigation extraction module and a U-turn row changing module. The row navigation extraction module includes an RTK module and a vision module. The RTK module is used to apply the composition information in the harvesting operation by using the carrot agronomic parameter information to realize the position recognition of the carrot. The vision module is used to identify the row by the information of the ridge ditch edge to realize the real-time operation path planning. The U-turn row changing module is used to plan the path for the machine track, the land head, and the edge of the operation plot during the operation process.

[0059] The carrot agronomic parameter configuration subsystem includes a variety parameter configuration module, a cultivation parameter configuration module, and an environment parameter configuration module. The variety parameter configuration module is used to obtain the variety parameter configuration information. The cultivation parameter configuration module is used to obtain the cultivation parameter configuration information. The environment parameter configuration module is used to obtain the environment parameter configuration information.

[0060] Specifically, the carrot unmanned harvesting system is described by taking the harvesting of white carrots as an example. Figure 2It is a kind of intelligent transformation schematic diagram of radish harvesting machine in radish unmanned harvesting system provided by the present application, as shown in Figure 2 The radish unmanned harvesting system provided by the embodiment of the present application is based on a clamping type radish harvesting machine, and the basic structure includes a double-row radish supporting device, a fruit cutting device, a crawler power device, a radish fruit conveying device, a horizontal conveyor belt, a driving control room, a radish supporting electric control switch, a front-rear power control lever, a harvesting operation control lever, a radish machine direction and a harvesting arm up-down adjustment, a navigation antenna, a millimeter wave radar, a laser radar, a binocular camera and a touch screen are arranged on the radish harvesting platform, the navigation antenna has a height interval of 1 meter, the touch screen provides an intelligent control entrance of the radish harvesting machine, multi-modal data of the millimeter wave radar, the laser radar and the binocular camera are fused to perform mapping and obstacle avoidance, all the control levers are connected to an electric control module of a rudder, including controlling the traveling direction and speed of the working platform and the lifting and lowering of the radish harvesting arm, and the radish harvesting machine is unmanned and improved based on the above.

[0061] The unmanned transformation of the radish harvesting equipment (i.e., the radish harvesting machine) provided by the embodiment of the present application integrates and applies electric control technology, Beidou navigation technology, computer vision recognition technology and control feedback technology in power control, operation tool control, operation path planning, accurate row alignment and fruit cutting of the radish harvesting equipment, and performs systematic adaptation.

[0062] Optionally, the unmanned movement control module includes an unmanned speed control module and an unmanned steering control module; the unmanned speed control module adopts a form of a rudder for voltage soft regulation; and the unmanned steering control module adopts proportional valve control.

[0063] Specifically, the unmanned speed control module adopts a form of a rudder for voltage soft regulation, so that the radish harvesting machine travels more stably, and the speed change process reduces oscillation.

[0064] The unmanned steering control module adopts proportional valve control, which improves the accuracy of steering control and facilitates flexible turning of the radish harvesting machine.

[0065] The unmanned steering control module mainly includes a remote control module, a wireless communication system, a Beidou satellite positioning system and an electric crawler chassis.

[0066] The radish harvesting machine uses a fusion recognition mode of a millimeter wave radar, a laser radar and a vision sensor to monitor obstacles in the traveling and reversing process in real time.

[0067] In the embodiment of the present application, in order to realize automatic control of the radish harvesting machine crawler chassis on the path during the radish harvesting process, it is necessary to define the pose during the traveling of the radish harvesting machine. Figure 3is a differential track tracking schematic diagram in a radish unmanned harvesting system provided by the present application, as shown in Figure 3 . The radish harvester track center is the motion coordinate base point, the track chassis center point O B (x, y), the speed size is , the left track speed size is , the right track speed size is , the direction of the O B point speed and the positive direction of the x-axis coordinate system is the heading angle of the track chassis, and the heading angle is positive in the counterclockwise direction of the x-axis coordinate positive direction. Figure 4 is a track type chassis power module schematic diagram in a radish unmanned harvesting system provided by the present application, and the left and right track speed directions or power directions of the radish harvester when turning left, straight and turning in place are shown in Figure 4 .

[0068] According to the principle of velocity equal to the derivative of displacement, the velocity is decomposed in x and y directions, and by the vector parallelogram decomposition rule, the kinematics equation of the track chassis can be obtained as follows:

[0069] (1)

[0070] (2)

[0071] Since the track chassis can be regarded as a rigid body structure, the motion of the left and right track center points at any instant during the motion can be regarded as circular motion around the same center, so the angular velocity of the left and right track center points is equal to the angular velocity of the track center point , and the angular velocity of the left and right track center points is:

[0072] (3)

[0073] wherein, represents the distance between the left and right track center points, which can be seen from Figure 3 .

[0074] By combining equations (1) to (3), the kinematics model of the track chassis can be obtained as follows:

[0075] (4)

[0076] For the radish unmanned harvesting operation, the planned path is usually composed of a series of points, and the path point contains the harvesting equipment posture, driving speed and other information. If the implement follows a point, it can be considered that the curve path following is achieved. The radish harvesting speed is usually below 3 km / h. For low-speed motion, it is appropriate to use a pure tracking algorithm to solve the turning radius. A coordinate system X B -O B -Y B (see Figure 3 ), the center point O B of the track model has coordinates (x, y), and the target point P B has coordinates (x b , y b ). In order to make the track chassis move from O B point to P B point, set the fixed differential speed of the left and right tracks of the track chassis to make O B point move to P B point with a circular arc trajectory. Set point O C as the center of the turning curve.

[0077] Define the angle between the straight line passing through the coordinate origin O B and the target point P B and the X B axis as , LE as the forward-looking distance of motion preview, R as the turning radius, and β as the circular arc angle turned by the track chassis from the initial position to the target point. According to the geometric relationship, we have:

[0078] (5)

[0079] We can get:

[0080] (6)

[0081] (7)

[0082] Project the target point P B onto the X B axis, and according to the geometric relationship, we can get:

[0083] (8)

[0084] (9)

[0085] Integrating equations (6) to (9), we get the turning radius of the track chassis:

[0086] (10)

[0087] Assuming the tracked chassis experiences pure friction with the ground during movement (i.e., slippage is not considered), then the speeds of the left and right tracks are... , With chassis center speed The following inherent relationship exists:

[0088] (11)

[0089] (12)

[0090] Substituting (10) into (11) and (12), the differential track speed setpoint can be obtained as follows:

[0091] (13)

[0092] Define x b Let α be the lateral error of path tracking, and α be the heading deviation of path tracking. When the target point's x-coordinate in the tracked chassis coordinate system... b The larger the absolute value, the greater the lateral distance of the harvester from the target path; the larger the absolute value of the angle between the target point and the x-axis of the tracked chassis coordinate system, the greater the heading error of the harvester from the target path.

[0093] The unmanned radish harvesting control subsystem provided in this embodiment of the invention improves the forward and backward movement and left and right turning devices of the radish harvester. It adopts a servo motor to solve the problems of abrupt switching of mechanical speed and unstable turning due to single-side track lock-up by using gentle voltage control. At the same time, electronic control is introduced to provide on-site remote control and remote algorithm control ports, so as to achieve effective control of the radish harvester.

[0094] The unmanned radish harvesting subsystem includes a ridge navigation extraction module and a turn-around module. The ridge navigation extraction module includes an RTK module and a vision module.

[0095] The RTK module mainly utilizes basic information from the radish cultivation process to apply the mapping information during harvesting operations, thereby enabling the location identification of radishes.

[0096] The vision module is mainly used for ridge alignment. By recognizing information about the edges of the furrows, it enables real-time planning of work paths, making the process more accurate.

[0097] Optionally, the vision module uses the Hough operator and the Otsu method with adaptive thresholding to binarize the acquired images, and processes the loss function based on the Huber algorithm to identify the center line of the ridge and the boundaries on both sides of the ridge during the radish harvesting process.

[0098] Specifically, when the radish is aligned with the ridge based on the vision module, Hough operator and adaptive threshold Otsu (OTSU) are used for binarization, and the loss function is processed based on Huber algorithm to identify the ridge center line and the boundary of the two sides of the ridge in the radish harvesting process, so as to realize accurate alignment with the ridge in the radish harvesting process.

[0099] Optionally, the U-turn and line-changing module uses an elite ant colony algorithm for path optimization.

[0100] Specifically, the U-turn and line-changing module mainly plans paths for the access road, the land head and the edge of the work plot during the operation, and the elite ant colony algorithm is used for path optimization in the embodiments of the present application.

[0101] Figure 5 is a radish path planning diagram in a radish unmanned harvesting system provided by the present application, as shown in Figure 5 .

[0102] Figure 6 is an elite ant colony algorithm structure diagram in a radish unmanned harvesting system provided by the present application, as shown in Figure 6 .

[0103] Based on any of the above embodiments, various operators, algorithms and computer vision models are used in the radish harvesting work in the operation path planning, navigation, obstacle recognition, alignment with the ridge, clamping and pulling, cutting the ribbon and radish quality detection process. Since the models for the agricultural production site scene service usually need to have high precision and lightness, the embodiments of the present application are designed in the aspects of path planning and target detection.

[0104] Taking the fruit position detection of the radish clamping and pulling link as an example, the elite ant colony algorithm is used for path optimization in the path planning aspect; in the target detection aspect, the target is accurately obtained through the segmentation algorithm so as to enable the machine to quickly feedback and operate.

[0105] Optionally, the radish harvesting unmanned operation subsystem includes a clamping and pulling visual recognition module; the clamping and pulling visual recognition module uses an instance segmentation algorithm to obtain radish fruits.

[0106] The segmentation network of the instance segmentation algorithm uses a Yolov8s-seg model as a 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; the segmentation network includes an inverted bottle neck search module.

[0107] Specifically, the radish harvesting unmanned operation subsystem further comprises a clamping and pulling visual recognition module and a radish fruit and leaf visual recognition module. The clamping and pulling visual recognition module is used to identify the radish clamping and pulling position, so as to accurately clamp and pull the radish, and the radish fruit and leaf visual recognition module is used to identify the position of the radish fruit and leaf, so as to accurately cut the radish leaf.

[0108] In order to balance the recognition accuracy, low delay and small parameters (i.e. lower network parameter amount), the Yolov8s-seg instance segmentation algorithm supporting images and videos is used as the backbone network in the clamping and pulling visual recognition module and the radish fruit and leaf visual recognition module, and improvements are made on this basis. Figure 7 is an improved Yolov8s-seg network structure diagram in the radish unmanned harvesting system provided by the present application, as Figure 7 shown.

[0109] In order to increase the lightness of the network, the backbone adopts MobileNetv4, introduces a flexible inverted bottle neck search module. In addition, the inverted bottle neck search module, ConvNext, feedforward network and an additional depth divisible variant are fused, the model search efficiency is improved through an optimized NAS formula, a novel distillation technology is introduced to improve the model accuracy, and the LSKA module is used to replace the SPPF module in the Yolov8s-seg model, so as to improve the small target detection performance in the operation process, thereby reducing the calculation complexity and the occupation of the display memory.

[0110] The radish harvesting unmanned operation subsystem provided by the embodiment of the present application can perform harvesting unmanned operation path planning, fuse the radish harvester itself and the position of the harvesting arm, and use the improved ant colony algorithm for iterative calculation. Before each operation, the starting point, the ending point and the lowest energy consumption path planning of the operation are considered, so as to realize the analysis of the fixed land lowest energy consumption operation mode. The radish clamping and pulling image recognition can also be performed. The improved Yolov8s-seg model is used to identify the position of the radish fruit in real time. The MobileNetv4 is used to replace the backbone network, the LSKA is used to replace the SPPF module, the model training process with less parameters and light weight is realized, the small target detection performance in the operation process is improved, and the calculation complexity and the occupation of the display memory are reduced.

[0111] In the radish unmanned harvesting system provided by the present application, the radish agricultural parameter configuration subsystem is a key link of the agricultural machinery-agricultural technology fusion. By inputting the agricultural parameters, the driving requirements of the harvester are provided, including a variety parameter configuration module, a cultivation parameter configuration module and an environment parameter configuration module.

[0112] Optionally, the variety parameter configuration information comprises one or more of the following information:

[0113] Planting radish varieties:

[0114] Fruit size;

[0115] Mature time;

[0116] Water requirement at different links;

[0117] Fertilizer requirement at different links;

[0118] Suitable temperature;

[0119] Suitable humidity.

[0120] Specifically, the variety parameter configuration module mainly inputs the information of the planted radish varieties, fruit size, mature time, water requirement at different links, fertilizer requirement, suitable temperature, and humidity.

[0121] Optionally, the cultivation parameter configuration information includes one or more of the following information:

[0122] Overall operation area configuration;

[0123] Planting ridge height;

[0124] Ridge spacing;

[0125] Row spacing;

[0126] Plant spacing;

[0127] Cultivation time.

[0128] Specifically, the cultivation parameters are the focus of the configuration process, including overall operation area configuration, planting ridge height, ridge spacing, row spacing, plant spacing, and cultivation time, which can be used as scene parameter input for unmanned operation of agricultural machinery. In terms of agricultural machinery parameters, wheelbase, agricultural machinery position, starting point, and operation mode are configured.

[0129] Optionally, the environmental parameter configuration information includes one or more of the following information:

[0130] Air temperature;

[0131] Air humidity;

[0132] Soil temperature;

[0133] Soil humidity;

[0134] Light;

[0135] Soil PH.

[0136] Specifically, the environmental parameter configuration module mainly configures the suitable air temperature, air humidity, soil temperature, soil humidity, light, and soil PH for the basic information of radish varieties.

[0137] Based on any of the above embodiments, the embodiment of the present application adopts a computer vision method to identify the ridge line of the work, and the identification model architecture refers to the improved Yolov8s-seg (see Figure 7 ) method, the data input and labeling of the model are radish ridge lines, in this scenario, the harvesting arm is on the right side, so it is necessary to identify the left ridge line after the ridge, that is, through real-time correction by vision to ensure the accuracy when the unmanned vehicle travels.

[0138] Optionally, the radish unmanned harvesting system further comprises an edge server and a cloud server.

[0139] The radish unmanned harvesting system adopts a cloud-edge-end intelligent service architecture; the radish harvesting unmanned control subsystem and the radish harvesting unmanned operation subsystem are realized through a terminal; big data processing is processed in the edge server and / or the cloud server.

[0140] Specifically, Figure 8 is a radish unmanned harvesting system provided by the present application, as shown in Figure 8 The radish unmanned harvesting system in the embodiment of the present application adopts a "cloud, edge and end" collaborative architecture for model training, data transmission and model identification. During the model training process, high computing power and high processing performance services are usually required. Therefore, in the "cloud", it is realized by the control node of the cloud part Kubernetes and the node running by KubeEdge of the edge part. The Kubernetes control node follows the original data model of the cloud part, maintains the original control and data flow unchanged, that is, the node running by KubeEdge presents a normal node on the Kubernetes. Kubernetes can manage the node running by KubeEdge as it manages normal nodes. KubeEdge realizes the sinking of Kubernetes cloud computing arrangement containerized application on the basis of the Kubernetes control node through the CloudCore of the cloud part and the EdgeCore of the edge part. After periodically collecting the training data on site, it is labeled and trained according to the demand, the trained model is generated, and the scheduling is performed.

[0141] The "edge" and "terminal" cooperation refers to taking KubeEdge as a management program running on an edge node, responsible for managing the resources of unmanned operation process load on the edge node, the running state and failure of the operation terminal, etc. KubeEdge provides the required computing resources for EdgeXFoundry services, and is responsible for managing the entire life cycle of the EdgeX Foundry terminal service. EdgeX Foundry can collect, filter, store and mine data of various Internet of Things terminal devices through the managed harvesting job microservice module, and can also issue instructions to various Internet of Things terminal devices through the managed microservice to control the terminal devices. Through edge computing, model computing power deployment is carried out for production scenarios, and scene-based fine-tuning is carried out for different tasks. Embedded terminals are connected through high-bandwidth and low-latency networks, including video monitoring, obstacle avoidance terminals, navigation terminals, etc. at various positions in the radish harvesting machine.

[0142] The cloud edge terminal intelligent service framework in the radish unmanned harvesting system provided by the embodiment of the application involves sensors, controllers, video collectors, mechanical pose controllers, etc. in the process of executing unmanned operation tasks, that is, data is collected at the terminal at the operation site, small-power data analysis and processing are completed locally, and compression is performed before data transmission, and finally high-energy consumption and large-parameter model training is set in the cloud service, through model scene adaptation scheduling, the computing power consumption is reduced to the greatest extent.

[0143] The radish unmanned harvesting system provided by the application comprises a radish harvesting unmanned control subsystem, a radish harvesting unmanned operation subsystem and a radish agricultural parameter configuration subsystem, which intelligently define and control the forward and backward movement, direction, speed and lifting of the radish harvesting arm of the radish harvesting machine, and fuse the shape, terrain and planting agriculture of the vegetable field to plan the optimal operation path. Computer vision algorithms are integrated in multiple links to accurately identify radish fruits and radish flags, and more intelligent, delicate and accurate operation of the unmanned operation harvester is realized. All models of the unmanned operation transformation adopt a cloud edge terminal processing architecture, which reduces the computing power consumption through model scene adaptation scheduling. Thus, manpower, material resources and time are saved, and the radish unmanned harvesting efficiency, quality, and green and intelligent level of radish unmanned harvesting are improved.

[0144] Those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary universal hardware platform, and of course can also be implemented by hardware, through the description of the above embodiments. Based on such understanding, the above technical solutions essentially or in other words the part of the prior art that makes a contribution can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, and the like, and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0145] It should be noted that, in this document, the terms "comprising", "containing", or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or apparatus that includes a list of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent in such a process, method, article, or apparatus. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element. In addition, it should be pointed out that the scope of the methods and apparatus in the embodiments of the present application is not limited to performing functions in the order shown or discussed, but can also include performing functions in a substantially simultaneous manner or in reverse order, for example, the described method can be performed in an order different from that described, and various steps can also be added, omitted, or combined. In addition, the features described with reference to certain examples can be combined in other examples.

[0146] In the embodiments of the present application, "determining B based on A" means that A is considered as a factor when determining B. It is not limited to "determining B based only on A", but also includes "determining B based on A and C", "determining B based on A, C and E", "determining C based on A, and determining B based on C further", and the like. In addition, it can also include A as a condition for determining B, for example, "when A meets the first condition, determining B using the first method"; for example, "when A meets the second condition, determining B"; for example, "when A meets the third condition, determining B based on the first parameter"; and the like. Of course, A can also be a condition for determining B, for example, "when A meets the first condition, determining C using the first method, and further determining B based on C".

[0147] In the present application, the term "multiple" means two or more, and other quantifiers are similar.

[0148] It should be pointed out finally that the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit the same; and although the present application has been described in detail with reference to the foregoing embodiments, it should be appreciated by those skilled in the art that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features thereof can be replaced equivalently; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A system for the automated harvesting of radish plants, characterized in that The system comprises a radish harvesting unmanned control subsystem, a radish harvesting unmanned operation subsystem, and a radish agronomic parameter configuration subsystem. The radish harvesting unmanned control subsystem comprises an unmanned movement control module and an unmanned radish harvesting module; the unmanned movement control module is mainly composed of a remote control module, a wireless communication system, a Beidou satellite positioning system, and an electric caterpillar chassis; the unmanned radish harvesting module uses an included angle sensor to control the height, and the included angle is used to deduce the height of the operation platform and the ridge. The radish harvesting unmanned operation subsystem comprises a ridge navigation extraction module and a U-turn and row changing module; the ridge navigation extraction module comprises an RTK module and a vision module, the RTK module is used to apply the layout information in the harvesting operation by using the radish agronomic parameter information, and the position recognition of the radish is realized; the vision module is used to navigate the ridge by recognizing the ridge ditch edge information, and realize real-time operation path planning; the U-turn and row changing module is used to plan the path to the machine plough way, the land head, and the edge of the operation land block during the operation. The radish agronomic parameter configuration subsystem comprises a variety parameter configuration module, a cultivation parameter configuration module, and an environment 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; and the environment parameter configuration module is used to obtain environment parameter configuration information. The radish harvesting unmanned operation subsystem comprises a clamping and pulling visual recognition module; the clamping and pulling visual recognition module uses an instance segmentation algorithm to obtain radish fruits. The segmentation network of the instance segmentation algorithm takes a Yolov8s-seg model as a 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 comprises a reverse bottle neck search module. The vision module uses a Hough operator and an adaptive threshold Otsu method to binarize the collected images respectively, and processes the loss function based on a Huber algorithm to identify the ridge center line and the ridge side boundary in the radish harvesting process.

2. A system for harvesting radish roots without human intervention according to claim 1, characterized in that, The unmanned movement control module comprises an unmanned speed control module and an unmanned steering control module; the unmanned speed control module adopts the form of a rudder to softly regulate and control the voltage; and the unmanned steering control module adopts a proportional valve control.

3. The system of claim 1, wherein, The U-turn and row changing module uses an elite ant colony algorithm to optimize the path.

4. The system of claim 1, wherein, The radish unmanned harvesting system further comprises an edge server and a cloud server. The radish unmanned harvesting system adopts a cloud edge intelligent service architecture; the radish harvesting unmanned control subsystem and the radish harvesting unmanned operation subsystem are realized through a terminal; and big data processing is processed in the edge server and / or the cloud server.

5. The system of claim 1, wherein, The cultivation parameter configuration information comprises one or more of the following information: overall operation area configuration; planting ridge height; ridge spacing; row spacing; plant spacing; cultivation time.

6. The system of claim 1, wherein, The variety parameter configuration information comprises one or more of the following information: planted radish variety: contains fruit size; Mature time; Water requirement of different links; Fertilizer requirement of different links; Suitable temperature; Suitable humidity.

7. The system of claim 1, wherein, The environmental parameter configuration information includes one or more of the following information: Air temperature; Air humidity; Soil temperature; Soil humidity; Illumination; Soil PH.

Images