A smart harvesting and sorting system and method

Abstract

This invention discloses an intelligent harvesting and sorting system and method, comprising: an intelligent harvesting and sorting mechanical gripper and a harvesting and sorting system. The harvesting and sorting system includes: a computer processing module and, electrically connected to the computer processing module, a dielectric spectrum generating electrode plate, a near-infrared light source generating device, a dielectric spectrum receiving plate, a fiber optic probe, a near-infrared spectrometer, an image acquisition device, and a power supply module; the fiber optic probe is connected to the near-infrared spectrometer. The intelligent harvesting and sorting mechanical gripper includes: a hydraulic mechanism and a clamping mechanism; the computer processing module is connected to the hydraulic mechanism, and after receiving control command information sent by the computer processing module, the hydraulic mechanism controls the clamping mechanism to move; the dielectric spectrum generating electrode plate, the near-infrared light source generating device, the dielectric spectrum receiving plate, and the fiber optic probe are respectively disposed on the four mechanical grippers of the clamping mechanism; this invention can adaptively adjust the clamping force on objects and automatically sort them by quality.

Description

Technical Field

[0001] This invention relates to the field of intelligent harvesting and sorting devices, and more specifically to an intelligent harvesting and sorting system and method. Background Technology

[0002] Robotic arms are a new type of device developed in the process of mechanized and automated production. Although robotic arms are not as dexterous as human hands, they have the characteristics of being able to continuously repeat work and labor, without fatigue, without fear of danger, and with greater strength in grasping and lifting heavy objects than human hands. They have been widely used in fields such as machinery manufacturing, metallurgy, electronics, light industry, and atomic energy.

[0003] With the integration of machine vision, image processing, and perception technologies with robotic arms, robotic arms are becoming increasingly intelligent, enabling them to perform more functions and showing good results in areas such as remote control, item sorting, and crop harvesting.

[0004] As a major fruit and vegetable producing country, China produces over 300 million tons of fruit and over 800 million tons of vegetables annually. Not only is the output large, but the variety is also abundant, with over 300 types of fruits cultivated, each containing different varieties. For example, there are over 17 varieties of common citrus and over 54 varieties of apples. The planting, harvesting, and sorting of these fruits require a large amount of manpower, and the repetitive movements over long periods can easily lead to occupational diseases and work-related injuries. The use of robotic arms reduces labor costs, ensures the health and safety of workers, and improves work efficiency and quality.

[0005] However, there are still several problems with using robotic arms for fruit picking and sorting: 1. Harvesting is done using rigid gripper arms. Since they lack the ability to detect force, they are generally set to a fixed gripping force. During the harvesting process, the gripping force cannot be dynamically adjusted according to the condition of the fruit. This often damages the skin or internal tissue of the fruit, reducing its shelf life. If the gripping force is insufficient, the fruit will slip and the harvesting will not be completed smoothly. 2. In conventional harvesting scenarios, the identification and localization of objects generally rely on machine vision, which uses deep learning to segment and identify objects. This method usually depends on powerful graphics cards as computing platforms, which is costly, has high requirements for lighting conditions, and the equipment structure is relatively complex and cannot be miniaturized. 3. Harvesting and sorting are two separate processes. Often, the harvested fruit is transported to the factory and then sorted according to its quality. This involves many steps and a large workload. 4. Existing robotic arms and other picking and sorting equipment are mostly specialized equipment that can only pick or sort a specific type of fruit or vegetable. For example, a robotic arm for picking apples cannot be used to pick tomatoes, and a robotic arm for sorting radishes cannot be used to sort eggs. This is mainly because the robotic arm cannot identify the material of the target object, cannot autonomously determine what the object being gripped is, and the gripping force of the clamp cannot be automatically adjusted according to the situation. This can result in either insufficient gripping or excessive force causing damage to the target. Furthermore, it cannot simultaneously test different fruits and vegetables, has poor versatility, and is limited in its usage. Summary of the Invention

[0006] The purpose of this invention is to provide an intelligent harvesting and sorting system and method. The intelligent harvesting and sorting system is applied to a robotic arm, which can identify the material and type of the object during the approach, gripping and harvesting process, and can judge the internal quality of the object. It is suitable for intelligent harvesting and sorting of various fruits and vegetables.

[0007] This addresses the shortcomings of the existing technologies.

[0008] To achieve the above objectives, the present invention provides the following technical solution: An intelligent harvesting and sorting system includes: an intelligent harvesting and sorting mechanical claw and a harvesting and sorting system. The harvesting and sorting system includes: a computer processing module and, electrically connected to the computer processing module, a dielectric spectrum generating electrode plate, a near-infrared light source generating device, a dielectric spectrum receiving plate, a fiber optic probe, a near-infrared spectrometer, an image acquisition device, and a power supply module. The fiber optic probe is connected to the near-infrared spectrometer, and the power supply module provides operating power to the intelligent harvesting and sorting system. The intelligent picking and sorting mechanical claw includes a hydraulic mechanism and a clamping mechanism; the computer processing module is connected to the hydraulic mechanism, and after receiving control command information sent by the computer processing module, the hydraulic mechanism controls the clamping mechanism to move; the clamping mechanism includes four clamping arms and four mechanical claws, and the ends of the four clamping arms are respectively connected to the four mechanical claws through a gear mechanism; the hydraulic mechanism is drivenly connected to the four clamping arms respectively. The dielectric spectrum generating electrode plate, the near-infrared light source generating device, the dielectric spectrum receiving plate, and the fiber optic probe are respectively mounted on four mechanical grippers. The dielectric spectrum generating electrode plate and the dielectric spectrum receiving plate are arranged opposite to each other, and the near-infrared light source generating device and the fiber optic probe are arranged opposite to each other. The image acquisition device is located in the middle of the four mechanical grippers.

[0009] The image acquisition device is used to acquire images of objects, identify and judge the position and shape of objects, perform rough positioning of large scenes, and guide the movement of intelligent picking and sorting robotic claws. The dielectric spectrum generating electrode plate is used to generate alternating electromotive force excitation. The dielectric spectrum receiving board is used to receive the response voltage signal generated by alternating electromotive force excitation, so as to realize the measurement of the approach distance of the intelligent picking and sorting mechanical claw to the object, the material of the object and the deformation. The near-infrared light source generator is used to emit near-infrared light signals; The fiber optic probe is used to receive near-infrared light signals emitted by the near-infrared light source generator. The near-infrared spectrometer is used to detect near-infrared light transmission signals, thereby enabling the measurement of the internal quality of an object; The computer processing module is used to calculate the capacitance and complex relative permittivity between the dielectric spectrum generating electrode plate and the dielectric spectrum receiving plate during the process of using the intelligent picking and sorting mechanical claw to clamp the object, and after data processing, obtain dielectric spectrum data information between the dielectric spectrum generating electrode plate and the dielectric spectrum receiving plate. Based on the detection of the size and change of capacitance, the intelligent picking and sorting robotic claw approaches the object at a certain distance and the shape of the object. The computer processing module sends control command information to adjust the precise position of the intelligent picking and sorting robotic claw and the position of the maximum outer diameter of the object. The system identifies the material and deformation of objects based on dielectric spectrum data, and determines the required clamping force based on the material and deformation of the objects; it also detects internal defects and analyzes nutritional components of objects based on near-infrared light transmission signals, and achieves automatic sorting by combining the object size detection results.

[0010] Furthermore, it also includes a temperature sensor; the temperature sensor is mounted on the intelligent picking and sorting mechanical claw and connected to the computer processing module, and is used to provide temperature compensation for the detection of near-infrared light and dielectric spectrum.

[0011] Furthermore, the four mechanical grippers include a first mechanical gripper, a second mechanical gripper, a third mechanical gripper, and a fourth mechanical gripper; the first mechanical gripper and the third mechanical gripper are arranged opposite to each other, and the second mechanical gripper and the fourth mechanical gripper are arranged opposite to each other; The dielectric spectrum generating electrode plate is disposed on the first mechanical gripper; The near-infrared light source generator is mounted on the second mechanical gripper; The dielectric spectrum receiving plate is mounted on the third mechanical gripper; The fiber optic probe is mounted on the fourth mechanical gripper.

[0012] Furthermore, both the dielectric spectrum generating electrode plate and the dielectric spectrum receiving plate include multiple electrode contacts; each electrode contact includes multiple symmetrical electrode matrices.

[0013] Furthermore, the light source for the near-infrared light signal is a halogen lamp.

[0014] Furthermore, the frequency range of the alternating electromotive force excitation is in the range of 1MHz-2500MHz The cycles change at equal intervals.

[0015] Furthermore, the dielectric spectrum generating electrode plate and the dielectric spectrum receiving plate are externally covered with conductive silicone.

[0016] This invention also provides an intelligent harvesting and sorting method, comprising the following steps: Initialize the intelligent picking and sorting robotic claw; An image acquisition device is used to acquire images of an object and locate its position. Calculate the capacitance between the dielectric spectrum generating electrode plate and the dielectric spectrum receiving plate and the complex relative permittivity between the dielectric spectrum generating electrode plate and the dielectric spectrum receiving plate, use the capacitance value to identify the shape of the object, and determine the clamping position; Data processing is performed on the capacitance value and complex relative permittivity to obtain dielectric spectrum data. The material and type of the object are obtained from the dielectric spectrum data, and the required clamping force of the object is determined based on the material and type of the object. The system uses near-infrared light transmission signals to determine the internal mass of objects and automatically sorts them.

[0017] Furthermore, the calculation of the capacitance generated between the first and third mechanical grippers specifically involves: The inductive capacitance C when there is no object being gripped between the first and third robotic grippers b for: C b =C 0 +C 1 in, C 0 =ε 0 S / D The capacitance generated by the air dielectric between the electrodes. ε 0 The dielectric constant of air is S The area of ​​the mechanical gripper. D The distance between the electrodes. C 1 This is the sum of the boundary capacitance and the distributed capacitance caused by the measurement leads and the measurement system; Capacitance when the first and third robotic claws are close to each other but not in contact with the object C d for: C d =Cs+C2 + C 1

[0018] C 2 =ε 0 S 1 / D ; in, The dielectric constant of the object, t For the thickness of the object, x The area of ​​the object; C 2 An air-plate capacitor covering an area where no other object is present, except for the object being held. S 1 The area between the plates other than the object being held. S 1 =S-x ; D Given the distance between the electrodes, we get: C S =C d -C b +C 0 -C 2 When the object is larger than the electrode on the robotic gripper C 2 It can be ignored; When the first and third robotic grippers contact the object, the first robotic gripper applies an alternating electromotive force (EMF), and the third robotic gripper receives the EMF signal. The complex relative permittivity between the first and third robotic grippers... ε* : ε*=ε'-jε''=ε+k / (jε 0 w) in, ε' yes ε* The real part, ε'' yes ε* The imaginary part, ε The relative permittivity, k For electrical conductivity, w It is angular frequency. w= 2Πf , f It's frequency. Represents the imaginary unit; After baseline correction, normalization, and smoothing and denoising, the capacitance and complex relative permittivity of the object after electromotive force excitation signal are subjected to cluster analysis to determine the material and type of the object.

[0019] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects: 1. This invention can detect the type and variety of clamped fruits and vegetables through dielectric spectrum analysis at different frequencies; 2. By using an infrared detection device on the mechanical gripper, the quality of fruits and vegetables can be detected and sorted quickly during the gripping and picking process; 3. Using dielectric spectrum analysis to identify fruit and vegetable types and varieties is applicable to most non-conductive fruits and vegetables with high moisture content, demonstrating good versatility; 4. Using a mechanical gripper to detect the capacitance between the grippers enables non-contact detection of the shape of fruits and vegetables. Pressure is sensed through capacitance and changes, and the gripping force is adaptively adjusted to avoid damage to fruits and vegetables during the gripping process. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0021] The present invention, utilizing an intelligent harvesting and sorting system and method, will be further described below with reference to the accompanying drawings; Figure 1 This is a structural diagram of the intelligent picking and sorting mechanical claw provided by the present invention; Figure 2 This is a schematic diagram of the dielectric spectrum generating electrode plate and dielectric spectrum receiving plate provided by the present invention; Figure 3 This is a schematic diagram of the structure of each electrode contact in the dielectric spectrum generating electrode plate and dielectric spectrum receiving plate of the intelligent picking and sorting system provided by the present invention; Figure 4 This is a schematic diagram of the intelligent harvesting and sorting method provided by the present invention; Figure 5 This is a schematic diagram of the capacitor circuit structure during the clamping process of the first and third mechanical claws provided by the present invention.

[0022] In the figure, 101-Dielectric spectrum receiving plate; 102-Dielectric spectrum generating electrode plate; 103-Fiber optic probe; 104-Infrared light source; 105-Sapphire glass; 106-Hydraulic mechanism; 107-Gear mechanism; 108-Image acquisition device; 110-Temperature sensor; 111-Computer processing module; 201-First mechanical gripper; 202-Third mechanical gripper; 203-Second mechanical gripper; 204-Fourth mechanical gripper. Detailed Implementation

[0023] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0024] To better understand the purpose, structure, and function of this invention, the invention will be described in further detail below with reference to the accompanying drawings.

[0025] like Figure 1 As shown, an intelligent harvesting and sorting system includes: an intelligent harvesting and sorting mechanical claw and a harvesting and sorting system. The harvesting and sorting system includes: a computer processing module 111 and a dielectric spectrum generating electrode plate 102, a near-infrared light source generating device 104, a dielectric spectrum receiving plate 101, an optical fiber probe 103, a near-infrared spectrometer, an image acquisition device 108, and a power module, all electrically connected to the computer processing module 111. The optical fiber probe 103 is connected to the near-infrared spectrometer, and the power module provides operating power to the intelligent harvesting and sorting system. The intelligent picking and sorting mechanical claw includes a hydraulic mechanism 106 and a clamping mechanism; the computer processing module 111 is connected to the hydraulic mechanism 106, and after receiving control command information sent by the computer processing module 111, the hydraulic mechanism 106 controls the clamping mechanism to move; the clamping mechanism includes four clamping arms and four mechanical claws, and the ends of the four clamping arms are respectively connected to the four mechanical claws through a gear mechanism 107; the hydraulic mechanism 106 is drivenly connected to the four clamping arms respectively; The dielectric spectrum generating electrode plate 102, the near-infrared light source generating device 104, the dielectric spectrum receiving plate 101, and the fiber optic probe 103 are respectively mounted on four mechanical grippers. The dielectric spectrum generating electrode plate 102 and the dielectric spectrum receiving plate 101 are arranged opposite to each other, and the near-infrared light source generating device 104 and the fiber optic probe 103 are arranged opposite to each other. The image acquisition device 108 is located in the middle of the four mechanical grippers.

[0026] The power drive module is connected to the dielectric spectrum generating electrode plate 102, the dielectric spectrum receiving plate 101, the near-infrared spectrometer, the near-infrared light source generating device 104, the image acquisition device 108 and the hydraulic mechanism 106 respectively, and is used to provide electrical power to drive each component. It should be noted that the clamping structure also includes four clamping arms, which are respectively connected to the first mechanical claw 201, the second mechanical claw 203, the third mechanical claw 202 and the fourth mechanical claw 204 via a gear mechanism 107; the hydraulic mechanism 106 is respectively connected to the four clamping arms.

[0027] It should be noted that: four mechanical claws are set up, of which the first mechanical claw 201 and the third mechanical claw 202 are equipped with dielectric spectrum generators and sensing devices, respectively, for the identification of fruit and vegetable types and varieties; the second mechanical claw 203 and the fourth mechanical claw 204 are equipped with near-infrared generators and detection devices, respectively, for the detection of fruit and vegetable quality; the image acquisition device is specifically: a camera, used for image acquisition and positioning of the basic position of the fruit and vegetables, the camera is set at the center of the four mechanical claws, and the camera is equipped with sapphire glass to protect the camera.

[0028] By continuously acquiring images during the coordinate changes of the robotic arm using a camera, the relative position of the robotic arm and the fruits and vegetables is determined. This camera only needs to distinguish between fruits and vegetables and leaves and their positions, without needing to identify the types of fruits and vegetables. Therefore, the requirements for camera pixels, lighting conditions, image processing speed, and computing power are low. After approaching the fruits and vegetables, the shape of the fruits and vegetables is identified by the change in the sensing capacitance between the first and third robotic claws. By calculating the maximum sensing capacitance, the maximum outer diameter of the fruits and vegetables is obtained to adjust the clamping position. After the robotic claws close and contact the fruits and vegetables, an alternating electromotive force excitation is applied to the first robotic claw 201, and the response signal of the sensing device of the third robotic claw 202 is measured. Through group analysis, the material and type of the fruits and vegetables being clamped are determined, and the required clamping force is determined based on the material and type of the fruits and vegetables. During the clamping process, the change in piezoelectric capacitance between the first and third mechanical claws determines the clamping force. The clamping force is adaptively adjusted according to the material and type of the fruits and vegetables obtained, so as to stabilize the clamping while avoiding damage to the skin or internal tissue of the fruits and vegetables. The second mechanical claw 203 generates a near-infrared light signal, and the fourth mechanical claw 204 is equipped with a near-infrared spectroscopy detection device to detect the near-infrared transmission signal. Through chemometric analysis, the internal quality of the clamped fruits and vegetables is determined.

[0029] The camera is fitted with sapphire glass 105 to prevent it from being scratched.

[0030] The image acquisition device 108 is used to acquire images of objects, identify and judge the position and shape of objects, perform rough positioning of large scenes, and guide the movement of intelligent picking and sorting robotic claws. The dielectric spectrum generating electrode plate 102 is used to generate alternating electromotive force excitation; The dielectric spectrum receiving board 101 is used to receive the response voltage signal generated by the alternating electromotive force excitation, so as to realize the intelligent picking and sorting mechanical claw to measure the approach distance of the object, the material of the object and the deformation. The near-infrared light source generator 104 is used to emit near-infrared light signals; The fiber optic probe is used to receive near-infrared light signals emitted by the near-infrared light source generator 104; The near-infrared spectrometer is used to detect near-infrared light transmission signals, thereby enabling the measurement of the internal quality of an object; The computer processing module 111 is used to calculate the capacitance and complex relative permittivity between the dielectric spectrum generating electrode plate 102 and the dielectric spectrum receiving plate 101 during the process of using the intelligent picking and sorting mechanical claw to clamp the object, and after data processing, obtain dielectric spectrum data information between the dielectric spectrum generating electrode plate 102 and the dielectric spectrum receiving plate 101. Based on the detection of the size and change of capacitance, the intelligent picking and sorting mechanical claw approaches the object at a distance and the shape of the object. The computer processing module 111 sends control command information to adjust the precise position of the intelligent picking and sorting mechanical claw and the position of the maximum outer diameter of the object. Based on dielectric spectrum data, the system identifies the material and deformation of objects and determines the required clamping force accordingly. It also detects internal defects such as rot, disease, and pests in fruits and vegetables based on near-infrared light transmission signals, analyzes nutritional components such as sugar content, moisture content, acidity, and vitamin content, and achieves automatic sorting by combining the results of fruit and vegetable size detection.

[0031] It should be noted that: dielectric spectrum is used to identify the types of fruits and vegetables and provide information on their internal density, while near-infrared spectroscopy can determine the chemical composition and nutritional status of the fruit based on light absorption feedback; by identifying the types of fruits and vegetables and their internal density through dielectric spectrum, the identification capability of infrared spectroscopy chemometrics analysis is improved, and problems such as rot, water core, hollowness, or insect infestation can be detected inside the fruits and vegetables; during the harvesting process, the quality of the fruits and vegetables inside the fruit and vegetables is automatically judged and automatically sorted. It also includes a temperature sensor 110; the temperature sensor 110 is mounted on the intelligent picking and sorting mechanical claw and connected to the computer processing module 111, and is used to provide temperature compensation for the detection of near-infrared light and dielectric spectrum.

[0032] The dielectric spectrum generating electrode plate 102 is disposed on the first mechanical claw 201 of the intelligent picking and sorting mechanical claw; The near-infrared light source generating device 104 is mounted on the second mechanical claw 203 of the intelligent picking and sorting mechanical claw; The dielectric spectrum receiving plate 101 is disposed on the second mechanical claw 203 of the intelligent picking and sorting mechanical claw; The fiber optic probe 103 is mounted on the fourth mechanical gripper 204.

[0033] like Figure 2 and Figure 3 As shown, both the dielectric spectrum generating electrode plate 102 and the dielectric spectrum receiving plate 101 include multiple electrode contacts 401; each electrode contact 401 includes multiple symmetrical electrode matrices 301.

[0034] It should be noted that: the external conductive silicone coating protects the electrodes and the clamped object while increasing friction; within the electrode contacts of the dielectric spectrum generating electrode plate, a symmetrical electrode matrix generates alternating electromotive force in a differential manner, eliminating electromagnetic interference from the external environment.

[0035] The light source for the near-infrared light signal is a halogen lamp.

[0036] The frequency range of the alternating electromotive force is in 1MHz-2500MHz The cycles change at equal intervals.

[0037] The dielectric spectrum generating electrode plate 102 and the dielectric spectrum receiving plate 101 are externally covered with conductive silicone 302.

[0038] like Figure 4 As shown, the present invention also provides an intelligent harvesting and sorting method, comprising the following steps: Initialize the intelligent picking and sorting robotic claw; The image acquisition device 108 is used to acquire images of the object and locate the position of the object. The capacitance between the dielectric spectrum generating electrode plate 102 and the dielectric spectrum receiving plate 101 and the complex relative permittivity between the dielectric spectrum generating electrode plate 102 and the dielectric spectrum receiving plate 101 are calculated. The shape of the object is identified by the capacitance value, and the clamping position is determined. Data processing is performed on the capacitance value and complex relative permittivity to obtain dielectric spectrum data. The material and type of the object are obtained using the dielectric spectrum data, and the required clamping force of the object is determined based on the material and type of the fruit and vegetable. By utilizing near-infrared light transmission signals and judging the material and type of the clamped object based on the dielectric spectrum, the algorithm of infrared spectral signals is enhanced, thereby judging the internal quality of the object and automatically sorting it.

[0039] The principle behind infrared spectroscopy for judging the internal quality of fruits and vegetables is that specific molecules in a substance have the property of absorbing specific wavelengths of near-infrared light. By measuring the absorption of infrared light at specific wavelengths, the composition and content of this substance can be analyzed.

[0040] The capacitance generated between the dielectric spectrum generating electrode plate 102 and the dielectric spectrum receiving plate 101 is calculated as follows: like Figure 5 As shown, the sensing capacitance C when there is no object being held between the first mechanical gripper 201 and the third mechanical gripper 202 is... b for: C b =C 0 +C 1 in, C 0 =ε 0 S / D The capacitance generated by the air dielectric between the electrodes. ε 0 The dielectric constant of air is S The area of ​​the mechanical gripper. D The distance between the electrodes. C 1 This is the sum of the boundary capacitance and the distributed capacitance caused by the measurement leads and the measurement system; The capacitance C when the first mechanical gripper 201 and the third mechanical gripper 202 are close to each other but not in contact with an object d for: C d = Cs+C 2 +C 1

[0041] in, The dielectric constant of the object, t For the thickness of the object, x The area of ​​the object; C 2 An air-plate capacitor covering an area where no other object is present, except for the object being held. S 1 The area between the plates other than the object being held. S 1 =S-x ; D Given the distance between the electrodes, we get: C S =C d -C b +C 0 -C 2 When the object is larger than the electrode on the robotic gripperC 2 The change in the object's shape can be obtained by varying the travel distance of the mechanical gripper, thus determining the position of the object's maximum outer diameter for clamping.

[0042] It should be noted that dielectric properties are related to the composition of the material, and changes in the composition of the material will be reflected in the dielectric properties. The dielectric spectrum shows the change of the dielectric constant in an electromagnetic field with different frequencies. This method contains more information than the dielectric constant method at a single frequency. Therefore, dielectric spectrum technology is a commonly used method for measuring non-conductive materials and semi-solid materials with high moisture content. Different types of materials can be identified based on the dielectric spectrum.

[0043] When the first mechanical gripper 201 and the third mechanical gripper 202 come into contact with the object, the first mechanical gripper 201 applies an alternating electromotive force, and the third mechanical gripper 202 receives the alternating electromotive force signal. The complex relative permittivity ɛ* between the first mechanical gripper 201 and the third mechanical gripper 202 is: ε*=ε'-jε''=ε+k / (jε 0 w) in, ε' yes ε* The real part, ε'' yes ε* The imaginary part, ε The relative permittivity, k For electrical conductivity, w It is angular frequency. w= 2Πf , f It's frequency. Represents the imaginary unit; The changes in material composition are reflected in the dielectric spectrum. Different materials and types of fruits and vegetables have different charge storage capabilities. The dielectric spectrum curve generated by alternating potential can reflect this difference. After baseline correction, normalization, smoothing and denoising, the response spectrum data of the electromotive force excitation signal is clustered using the group analysis method to determine the material and type of fruits and vegetables.

[0044] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An intelligent harvesting and sorting system, characterized in that, include: The intelligent picking and sorting mechanical claw and picking and sorting system include: a computer processing module (111) and a dielectric spectrum generating electrode plate (102), a near-infrared light source generating device (104), a dielectric spectrum receiving plate (101), an optical fiber probe (103), a near-infrared spectrometer, an image acquisition device (108), and a power supply module, all electrically connected to the computer processing module (111); the optical fiber probe (103) is connected to the near-infrared spectrometer, and the power supply module is used to provide working power for the intelligent picking and sorting system; The intelligent picking and sorting mechanical claw includes a hydraulic mechanism (106) and a clamping mechanism; the computer processing module (111) is connected to the hydraulic mechanism (106), and the hydraulic mechanism (106) controls the clamping mechanism to move after receiving the control command information sent by the computer processing module (111); the clamping mechanism includes four clamping arms and four mechanical claws, and the ends of the four clamping arms are respectively connected to the four mechanical claws through a gear mechanism (107); the hydraulic mechanism (106) is driven to connect to the four clamping arms respectively; The dielectric spectrum generating electrode plate (102), the near-infrared light source generating device (104), the dielectric spectrum receiving plate (101), and the fiber optic probe (103) are respectively mounted on four mechanical claws. The dielectric spectrum generating electrode plate (102) and the dielectric spectrum receiving plate (101) are arranged opposite to each other, and the near-infrared light source generating device (104) and the fiber optic probe (103) are arranged opposite to each other. The image acquisition device (108) is located in the middle of the four mechanical grippers; The image acquisition device (108) is used to acquire images of objects, identify and judge the position and shape of objects, perform rough positioning of large scenes, and guide the operation of intelligent picking and sorting mechanical claws. The dielectric spectrum generating electrode plate (102) is used to generate alternating electromotive force excitation; The dielectric spectrum receiving plate (101) is used to receive the response voltage signal generated by the alternating electromotive force excitation, so as to realize the measurement of the approach distance of the intelligent picking and sorting mechanical claw to the object, the material of the object and the deformation. The near-infrared light source generator (104) is used to emit near-infrared light signals; The fiber optic probe is used to receive near-infrared light signals emitted by the near-infrared light source generator (104); The near-infrared spectrometer is used to detect near-infrared light transmission signals, thereby enabling the measurement of the internal quality of an object; The computer processing module (111) is used to calculate the capacitance and complex relative permittivity between the dielectric spectrum generating electrode plate (102) and the dielectric spectrum receiving plate (101) during the process of clamping the object using the intelligent picking and sorting mechanical claw, and after data processing, obtain dielectric spectrum data information between the dielectric spectrum generating electrode plate (102) and the dielectric spectrum receiving plate (101). Based on the detection of the size and change of capacitance, the intelligent picking and sorting mechanical claw approaches the object and the shape of the object. The computer processing module (111) sends control command information to adjust the precise position of the intelligent picking and sorting mechanical claw and the position of the maximum outer diameter of the object. Based on dielectric spectrum data, the material and deformation of the object are identified, and the required clamping force is determined according to the material and deformation of the object; in addition, the internal defects and nutritional components of the object are detected based on near-infrared light transmission signals, and automatic sorting is achieved by combining the object size detection results. It also includes a temperature sensor (110); the temperature sensor (110) is mounted on the intelligent picking and sorting mechanical claw and connected to the computer processing module (111) to provide temperature compensation for the detection of near-infrared light and dielectric spectrum; The four mechanical grippers include a first mechanical gripper (201), a second mechanical gripper (203), a third mechanical gripper (202), and a fourth mechanical gripper (204); the first mechanical gripper (201) and the third mechanical gripper (202) are arranged opposite to each other, and the second mechanical gripper (203) and the fourth mechanical gripper (204) are arranged opposite to each other; The dielectric spectrum generating electrode plate (102) is disposed on the first mechanical claw (201); The near-infrared light source generating device (104) is mounted on the second mechanical gripper (203); The dielectric spectrum receiving plate (101) is mounted on the third mechanical gripper (202); The fiber optic probe (103) is mounted on the fourth mechanical gripper (204); Both the dielectric spectrum generating electrode plate (102) and the dielectric spectrum receiving plate (101) include multiple electrode contacts (401); each electrode contact (401) includes multiple symmetrical electrode matrices (301).

2. The intelligent harvesting and sorting system according to claim 1, characterized in that, The light source for the near-infrared light signal is a halogen lamp.

3. The intelligent harvesting and sorting system according to claim 1, characterized in that, The frequency range of the alternating electromotive force excitation is cyclically and equally spaced between 1MHz and 2500MHz.

4. The intelligent harvesting and sorting system according to claim 1, characterized in that, The dielectric spectrum generating electrode plate (102) and dielectric spectrum receiving plate (101) are covered with conductive silicone (302).

5. An intelligent harvesting and sorting method, applied to the intelligent harvesting and sorting system according to any one of claims 1-4, characterized in that, Includes the following steps: Initialize the intelligent picking and sorting robotic claw; The image acquisition device (108) is used to acquire an image of the object and locate the position of the object; Calculate the capacitance between the dielectric spectrum generating electrode plate (102) and the dielectric spectrum receiving plate (101) and the complex relative permittivity between the dielectric spectrum generating electrode plate (102) and the dielectric spectrum receiving plate (101), use the capacitance to identify the shape of the object, and determine the clamping position; Data processing is performed on the capacitance value and complex relative permittivity to obtain dielectric spectrum data. The material and type of the object are obtained from the dielectric spectrum data, and the required clamping force of the object is determined based on the material and type of the object. The system uses near-infrared light transmission signals to determine the internal mass of objects and automatically sorts them.

6. The intelligent harvesting and sorting method according to claim 5, characterized in that, The capacitance generated between the dielectric spectrum generating electrode plate (102) and the dielectric spectrum receiving plate (101) is calculated as follows: The inductive capacitance C when there is no object being held between the first mechanical gripper (201) and the third mechanical gripper (202) b for: C b =C 0 +C 1 ； in, C 0 =ɛ 0 S / D The capacitance generated by the air dielectric between the electrodes. ɛ 0 The dielectric constant of air is S The area of ​​the mechanical gripper. D The distance between the electrodes. C 1 This is the sum of the boundary capacitance and the distributed capacitance caused by the measurement leads and the measurement system; Capacitance when the first and third robotic claws are close to each other but not in contact with the object C d for: C d =Cs+C 2 +C 1 ; ； ； in, The dielectric constant of the object, t For the thickness of the object, x The area of ​​the object; C 2 An air-plate capacitor covering an area where no other object is present, except for the object being held. S 1 The area between the plates other than the object being held. S 1 =Sx ; D Given the distance between the electrodes, we get: C S =C d -C b +C 0 -C 2 ; When the object is larger than the electrode on the robotic gripper C 2 It can be ignored; When the first and third robotic grippers contact the object, the first robotic gripper applies an alternating electromotive force (EMF), and the third robotic gripper receives the EMF signal. The complex relative permittivity between the first and third robotic grippers... ɛ* : ɛ*=ɛ'-j ɛ”=ɛ+k / (jɛ 0 w) ; in, ɛ' yes ɛ* The real part, ɛ” yes ɛ* The imaginary part, ɛ The relative permittivity, k For electrical conductivity, w It is angular frequency. w=2Π f , f It's frequency. Represents the imaginary unit; After baseline correction, normalization, and smoothing and denoising, the capacitance and complex relative permittivity of the object after electromotive force excitation signal are subjected to cluster analysis to determine the material and type of the object.

Images