Intelligent mango harvesting device

Abstract

The invention discloses an intelligent mango harvesting device, belongs to the field of fruit harvesting equipment, and aims to solve the technical problems of fruit damage and inaccurate maturity judgment caused by contact detection of existing mango harvesting equipment. According to the technical scheme, the method is characterized in that a laser emission module emits a pulse laser beam with the wavelength of 10.3-10.6 microns, and the pulse laser beam is focused by a lens group and then irradiated to a detection area which is 5-15 cm away from the surface of the mango; the laser interacts with ethylene molecules to generate a photoacoustic signal, and the photoacoustic signal is received by a microphone with a sound insulation cover; the signal processing unit converts the photoacoustic signal into a digital electric signal, and the controller calculates the ethylene concentration according to the digital electric signal and compares the ethylene concentration with a preset threshold to judge the maturity. The device is mainly used for automatic lossless harvesting in a mango garden, and the fruit damage rate is remarkably reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural machinery, and more particularly to an intelligent mango harvesting device. Background Art

[0002] Mangoes are an important tropical fruit grown worldwide. Traditionally, mango harvesting relies primarily on manual labor, which is inefficient and labor-intensive. Furthermore, mangoes can be easily damaged by collisions due to negligence during the picking process, affecting fruit quality. Furthermore, manual assessment of mango ripeness is often inaccurate, leading to premature or late harvesting and reducing the commercial value of the fruit. However, some fruit harvesting equipment is currently available on the market.

[0003] Existing mango harvesting equipment has the following defects: traditional contact ethylene sensors need to directly contact the fruit surface, which can easily lead to detection deviations due to residual pollutants on the sensor surface (such as pesticides and fruit powder), and the contact process may scratch the fruit skin, affecting the fruit quality. Summary of the Invention

[0004] One purpose of the present invention is to provide an intelligent mango harvesting device that uses a non-contact ethylene sensor and photoacoustic spectroscopy technology to detect ethylene concentration at a distance of 5-15 cm from the fruit. This avoids the risk of contamination and damage caused by contact between the sensor and the fruit, solves the problems of inaccurate detection and fruit damage caused by contact during maturity detection, and can quickly obtain ethylene content data.

[0005] Changes in ambient temperature and humidity will affect ethylene concentration detection, resulting in inaccurate maturity judgment.

[0006] The existing picking mechanism's claws are mostly rigid structures and lack force feedback. They cannot adaptively adjust the clamping force for mangoes of different maturity levels, which can easily cause damage.

[0007] During the fruit transportation and collection process, the fruit is easily damaged due to collision, and the efficiency will be affected if the collection container is not replaced in time.

[0008] The surface curvature of mango fruit is different, and traditional detection methods cannot adjust the detection distance in real time, affecting the detection accuracy.

[0009] Leaf reflections, water mist, etc. in the orchard environment will interfere with the scanning of the fruit surface and affect the accuracy of the three-dimensional contour data.

[0010] Changes in ambient light intensity can affect the structured light projection and lidar sampling effects, resulting in a low signal-to-noise ratio of the scanning data.

[0011] Changes in ambient humidity can affect structured light transmission and lead to scanning data errors.

[0012] Ambient temperature changes can cause structured light power drift, affecting scanning accuracy.

[0013] Environmental noise can interfere with photoacoustic signal reception, resulting in inaccurate ethylene concentration detection.

[0014] To achieve these objectives and other advantages of the present invention, an intelligent mango harvesting device is provided, comprising a traveling mechanism, a gripping mechanism disposed on the traveling mechanism, a conveying and collecting mechanism disposed below the gripping mechanism, and an ethylene content detection module. The ethylene content detection module includes a non-contact ethylene sensor. The non-contact ethylene sensor includes: A laser emission module, which is used to emit a pulsed laser beam to the detection area. The emission wavelength of the laser emission module is 10.3-10.6 μm; A lens assembly is provided on the outgoing optical path of the laser emission module. The lens assembly is used to focus the pulsed laser beam into a preset spot size and irradiate the air 5-15 cm from the surface of the mango fruit to form an ethylene molecule detection area. In the detection area, the pulsed laser beam interacts with the ethylene molecules released by the mango to generate a photoacoustic signal. A photoacoustic signal receiving unit, comprising a soundproof enclosure and a microphone, wherein the soundproof enclosure is used to suppress environmental noise interference, and the microphone is used to receive the photoacoustic signal generated in the detection area; a signal processing unit electrically connected to the microphone and configured to process the received photoacoustic signal to generate a digital electrical signal; The controller is connected to the signal processing unit through a data bus. The controller receives the digital electrical signal and converts the digital electrical signal into an ethylene gas concentration value, and compares the ethylene gas concentration value with a preset ethylene concentration threshold to determine the maturity of the mango.

[0015] Preferably, an environmental monitoring module is further included, the environmental monitoring module is connected to the controller, and the environmental monitoring module is equipped with a temperature and humidity sensor. The controller corrects the influence of the environment on the ethylene concentration detection through a temperature and humidity compensation algorithm. The specific formula is: C 修正 = C 实测 ×(1+ k 1Δ T + k 2Δ RH ) in, k 1. k 2 are the temperature compensation coefficient and humidity compensation coefficient, respectively, obtained through laboratory calibration, Δ RH is the difference between the current humidity and the standard humidity, Δ T is the difference between the current temperature and the standard temperature, C 实测is the original detection value of ethylene concentration directly obtained by the photoacoustic signal receiving unit, C 修正 is the final concentration value after adjustment based on the temperature and humidity compensation algorithm.

[0016] Preferably, the gripping mechanism comprises: Two clamping jaws, which are made of silicone material, and the inner wall of the clamping jaws is provided with a wave-shaped buffer groove with a depth of 0.3-0.5cm and a spacing of 1cm; A drive assembly includes a stepper motor connected to a controller, a bidirectional screw connected to the stepper motor, and two bases provided on the bidirectional screw. The roots of the two clamping jaws are respectively connected to the two bases so that the clamping force between the two clamping jaws can be adjusted by rotating the bidirectional screw. A force sensor is embedded in the root of the clamping jaw and connected to the controller. The force sensor is used to monitor the clamping force in real time and feed it back to the controller. When the clamping force monitored by the force sensor is ≥ a first preset threshold, the controller drives the stepper motor to open in the opposite direction so that the clamping force is less than the first preset threshold. When the clamping force monitored by the force sensor is ≤ a second preset threshold, the controller drives the stepper motor to rotate in the same direction so that the clamping force is greater than the second preset threshold, wherein the first preset threshold is greater than the second preset threshold.

[0017] Preferably, the conveying and collecting mechanism comprises: The inner wall of the delivery pipeline shall be paved with a silicone pad with a thickness of ≥5mm; A collection container is placed on a support base, wherein a weight sensor is provided on the support base, the weight sensor is connected to the controller, and the weight sensor is used to monitor the total weight of the mangoes in the collection container in real time; The prompt module is connected to the controller. When the weight sensor detects that the weight is ≥ the preset weight threshold, the controller sends a replacement signal to the prompt module. The prompt module includes at least one of the following: an audible and visual alarm, and a wireless communication module. The wireless communication module is used to send a replacement instruction to an external device.

[0018] Preferably, it also includes: A linear displacement platform is connected to the lens assembly, the linear displacement platform is driven by a telescopic motor, and the telescopic motor is connected to the controller; A three-dimensional profile sensor is located on the side of the laser emission module and is used to scan the surface curvature of the mango fruit in real time and generate surface concave and convex feature data; The dynamic focusing algorithm module is built into the controller and calculates the optimal focusing distance of the detection area based on the surface concave and convex feature data, which satisfies: When the curvature radius of the fruit surface is less than 50 mm, shorten the detection distance to 5-8 cm to enhance the laser energy density; When the curvature radius is greater than 150 mm, the detection distance is extended to 12-15 cm to expand the coverage of ethylene molecules; The linear displacement platform drives the lens group to move along the optical axis according to the controller instructions, so that the detection area matches the curvature radius of the fruit surface in real time, and the distance adjustment response time is less than 100ms.

[0019] Preferably, the three-dimensional profile sensor comprises: A dual-band laser scanning head, which includes a 940nm wavelength structured light projector and a 1550nm wavelength laser radar. The structured light projector is used to generate phase-coded stripes on the mango surface, and the laser radar is used to obtain millimeter-level depth point clouds on the fruit surface. Polarization filter set, which is set in front of the sensor receiving end, with its transmission axis direction orthogonal to the polarization direction of the scattered light in the orchard environment, to suppress leaf reflection and water mist interference; A dynamic calibration module, built into the sensor housing, adjusts the structured light projection power and lidar sampling frequency based on ambient light intensity to maintain a signal-to-noise ratio above 20dB. The controller fuses the phase data of the phase-coded stripes with the lidar point cloud data and reconstructs the fruit surface geometric model through a curvature compensation algorithm.

[0020] Preferably, the dynamic calibration module includes: The light intensity grading unit is configured as follows: When the ambient light intensity is less than 10,000 lux, the structured light projection power is controlled to 50 mW and the lidar sampling frequency is 10 Hz; When the illumination intensity is 10,000 lux or less than 80,000 lux, the structured light power is dynamically adjusted according to P = 50 + n × (I - 10,000), and the lidar sampling frequency is adjusted to 20 Hz, where I is the ambient light intensity, P is the structured light projection power, and n is 0.00286 mW / lux. When the light intensity is ≥80,000 lux, the structured light projection power is adjusted to 250 mW, the lidar anti-interference mode is activated and the sampling frequency is adjusted to 50 Hz.

[0021] Preferably, the dynamic calibration module further includes a multi-parameter coupling compensation unit, and the configuration of the multi-parameter coupling compensation unit is: The segmented humidity compensation module divides the structured light power additional humidity compensation coefficient into three levels of compensation strategies according to the ambient humidity data: Dry mode, ambient humidity RH < 65%: humidity compensation coefficient k3 = 1.0; Normal mode, ambient humidity 65% ​​≤ RH ≤ 90%: k3 = 1 + 0.03 × (RH - 65); High humidity mode, ambient humidity is RH>90%: k3=1.75+0.12×(RH-90).

[0022] Preferably, the multi-parameter coupling compensation unit is further configured as: a temperature-power coupling module, which adds a temperature compensation coefficient k4=1-0.015×(T-25) to the structured light power according to the ambient temperature T, and dynamically reduces the power when T>25℃ to suppress temperature drift.

[0023] Preferably, the composite soundproof enclosure comprises, from the inside to the outside: Sound absorbing layer, which is made of gradient density porous foam material, with density changing from 800 kg / m³ in the inner layer to 200 kg / m³ in the outer layer, and thickness of 10-15 mm; Damping layer, which is composed of butyl rubber and carbon fiber woven mesh, with a thickness of 2-3 mm; A reflective layer, comprising an aluminum alloy plate with a microprism array on its surface; Directional microphone array, which consists of four MEMS microphones arranged in a circular pattern at equal intervals in the center of the soundproof enclosure. Its directivity meets the following requirements: main lobe width ≤ 30°, focused on the center of the detection area, and sidelobe suppression ratio > 20 dB to reduce lateral noise interference; The adaptive noise cancellation module is embedded in the signal processing unit and uses the LMS filtering algorithm to extract the ambient noise reference signal in real time and cancel the common-mode noise component from the microphone array output signal.

[0024] The present invention has at least the following beneficial effects: First, the present invention realizes non-contact detection of ethylene content, avoiding contamination and damage that may be caused by contact between the sensor and the fruit. At the same time, it can quickly and accurately obtain ethylene content data, improving the accuracy and reliability of maturity judgment; through the temperature and humidity compensation algorithm, the accuracy of ethylene concentration detection under different environmental conditions is improved, making maturity judgment more reliable and ensuring accurate detection of ethylene concentration even in the complex environment of the orchard.

[0025] Second, the present invention realizes adaptive gripping of mangoes of different maturity through flexible grippers and a force feedback system, avoiding damage to the fruit caused by excessive gripping force, while ensuring gripping stability and reducing the damage rate of mangoes during the picking process; it realizes automatic collection of mangoes and container replacement prompt functions, improves collection efficiency, and reduces collision damage to the fruit during transportation, ensuring the quality of the fruit.

[0026] Third, the present invention uses a dynamic focusing algorithm and a linear displacement platform to enable the detection area to match the curvature of the fruit surface in real time, improving the accuracy and stability of ethylene concentration detection and ensuring accurate detection of ethylene content on mango surfaces of different shapes. Through a dual-band laser scanning head and a polarizing filter set, etc., precise scanning of the geometric features of the mango fruit surface is achieved, providing accurate data support for dynamic focusing, improving scanning accuracy and anti-interference capabilities, and ensuring accurate acquisition of fruit surface data under different environmental conditions.

[0027] Other advantages, objectives and features of the present invention will be reflected in part from the following description and will be understood by those skilled in the art through study and practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a structural schematic diagram of the intelligent mango harvesting device according to one of the technical solutions of the present invention; Figure 2 This is a schematic structural diagram of the clamping jaws described in one of the technical solutions of the present invention. DETAILED DESCRIPTION

[0029] The present invention will be described in further detail below in conjunction with the accompanying drawings so that those skilled in the art can implement the invention with reference to the description.

[0030] like Figure 1-2 As shown, the present invention provides an intelligent mango harvesting device, comprising a traveling mechanism 1, a gripping mechanism 5 provided on the traveling mechanism 1, a conveying and collecting mechanism provided below the gripping mechanism 5, and an ethylene content detection module. The ethylene content detection module includes a non-contact ethylene sensor. The non-contact ethylene sensor includes: Laser emission module (DFB laser), which is used to emit a pulsed laser beam to the detection area. The emission wavelength of the laser emission module is 10.3-10.6μm. The laser emission module emits a pulsed laser beam with a wavelength of 10.3-10.6μm. The common wavelength selection is 10.53μm, which corresponds to the absorption peak of ethylene molecules; A lens assembly is disposed on the outgoing optical path of the laser emission module (the outgoing optical path includes the complete path of the laser beam from the light source through the lens assembly to the detection area). The lens assembly is used to focus the pulsed laser beam to a preset spot size and irradiate the air 5-15 cm from the surface of the mango fruit, forming an ethylene molecule detection area. Within the detection area, the pulsed laser beam interacts with the ethylene molecules released by the mango to generate a photoacoustic signal. The lens assembly is disposed on the outgoing optical path of the laser emission module and can focus the pulsed laser beam into a spot with a diameter not exceeding 1 cm, irradiating the air 5-15 cm from the surface of the mango fruit, for example, at a distance of 8 cm, 10 cm, or 12 cm. A photoacoustic signal receiving unit, comprising a soundproof enclosure and a microphone. The soundproof enclosure is used to suppress environmental noise interference. The inner wall of the soundproof enclosure may be covered with sound-absorbing cotton to suppress environmental noise. The microphone is used to receive the photoacoustic signal generated in the detection area. a signal processing unit electrically connected to the microphone and configured to process the received photoacoustic signal (amplify, filter, digitize, etc. the signal) to generate a digital electrical signal; The controller is connected to the signal processing unit through a data bus. The controller receives digital electrical signals and converts the digital electrical signals into ethylene gas concentration values, and compares the ethylene gas concentration values ​​with the preset ethylene concentration threshold value to determine the maturity of the mango. The lens group can be a convex lens combination, the microphone can be a high-sensitivity electret microphone, the signal processing unit can be an integrated operational amplifier circuit, and the controller can be a single-chip microcomputer or an embedded processor; the soundproof cover can be made of metal or plastic, and the sound-absorbing cotton can be made of glass fiber or polyester fiber. During assembly, the laser emission module and the lens group are installed on an adjustable bracket to ensure that they can be accurately aligned with the detection area. The photoacoustic signal receiving unit is installed close to the detection area, and the controller is installed in the control box of the harvesting device; The traveling mechanism is also equipped with a solar power supply system, which is used to provide electricity to related electrical equipment; The working process is as follows: the laser emission module emits a pulsed laser, which is focused by the lens group and then irradiated onto the detection area. The laser interacts with the ethylene molecules released by the mango to generate a photoacoustic signal. The microphone receives the signal and transmits it to the signal processing unit. The signal processing unit processes the signal to generate a digital electrical signal. The controller receives the electrical signal and converts it into an ethylene concentration value. The value is then compared with a preset threshold to determine whether the mango is ripe. The non-contact ethylene sensor can be manually moved to detect the maturity of the fruit, or the non-contact sensor can be placed at the front end of the grasping mechanism 5. When the mango is detected to be ripe, the grasping mechanism 5 will grasp the mango. This technical solution enables non-contact detection of ethylene content, avoiding possible contamination and damage caused by contact between the sensor and the fruit. At the same time, it can quickly and accurately obtain ethylene content data, improving the accuracy and reliability of maturity judgment.

[0031] In another technical solution, an environmental monitoring module is also included. The environmental monitoring module is connected to the controller and is equipped with a temperature and humidity sensor. The controller corrects the influence of the environment on the ethylene concentration detection through a temperature and humidity compensation algorithm. The specific formula is: C 修正 = C 实测 ×(1+ k 1Δ T + k 2Δ RH ) in, k 1. k 2 are the temperature compensation coefficient and humidity compensation coefficient, respectively, obtained through laboratory calibration, Δ RH is the difference between the current humidity and the standard humidity, Δ T is the difference between the current temperature and the standard temperature, C 实测 is the original detection value of ethylene concentration directly obtained by the photoacoustic signal receiving unit, C 修正 is the final concentration value after adjustment based on the temperature and humidity compensation algorithm; The environmental monitoring module is equipped with a temperature and humidity sensor with an accuracy of ±2% RH and ±0.5°C, capable of collecting real-time temperature and humidity data. k1 and k2 are temperature and humidity compensation coefficients obtained through laboratory calibration. For example, in a laboratory, the temperature range is set to 15-40°C at 5°C intervals, and the humidity range is set to 40%-90% RH at 10% RH intervals. Regression analysis is performed on the ethylene detection data under each temperature and humidity combination to determine the values ​​of k1 and k2. ΔRH is the difference between the current humidity and the standard humidity, which can be set to 65% RH. ΔT is the difference between the current temperature and the standard temperature, which can be set to 25°C. The temperature and humidity sensor can be a commonly used capacitive or resistive sensor. The temperature and humidity sensor is installed near the ethylene detection module to ensure accurate collection of ambient temperature and humidity data in the detection area. The working process is as follows: the temperature and humidity sensor collects ambient temperature and humidity data in real time and transmits it to the controller. The controller uses the temperature and humidity compensation algorithm to correct the measured ethylene concentration based on the difference between the current temperature and humidity and the standard temperature and humidity, and obtains the final ethylene concentration correction value, thereby reducing the impact of ambient temperature and humidity on ethylene concentration detection; This technical solution improves the accuracy of ethylene concentration detection under different environmental conditions through a temperature and humidity compensation algorithm, making maturity judgment more reliable and ensuring accurate detection of ethylene concentration even in the complex environment of an orchard.

[0032] In another technical solution, the gripping mechanism 5 (the gripping mechanism 5 is connected to the robotic arm 4, and the robotic arm 4 is provided on the walking mechanism 1) includes: Two clamping jaws 505, which are made of silicone material, and the inner wall of the clamping jaws 505 is provided with a wave-shaped buffer groove with a depth of 0.3-0.5 cm and a spacing of 1 cm (the clamping jaws 505 include a housing and a clamping surface, and the non-contact ethylene sensor is embedded in the housing); A driving assembly includes a stepping motor 501 connected to a controller, a bidirectional lead screw 503 connected to the stepping motor 501, and two bases 502 provided on the bidirectional lead screw 503 (the two bases 502 rotate with the bidirectional lead screw 503 to achieve relative or back-to-back movement, a fixed seat is provided at the end of the mechanical wall, two positioning plates 504 are provided on the fixed seat, the bidirectional lead screw is rotatably provided between the two positioning plates 504, one end of which rotates through the positioning plate 504 and is connected to the stepping motor 501, a through hole is provided on the base 502, and limiting rods are provided in the two through holes, both ends of the limiting rods are connected to the two positioning plates 504, and the diameter of the through hole is larger than the diameter of the limiting rod), and the roots of the two clamping jaws 505 are respectively connected to the two bases 502 to adjust the clamping force between the two clamping jaws 505 by rotating the bidirectional lead screw 503; A force sensor is embedded in the base of the clamping jaw 505 and connected to the controller. The force sensor is used to monitor the clamping force in real time and provide feedback to the controller. When the clamping force monitored by the force sensor is ≥ a first preset threshold value (2N), the controller drives the stepper motor 501 to open in the opposite direction so that the clamping force is less than the first preset threshold value. When the clamping force monitored by the force sensor is ≤ a second preset threshold value (0.5N), the controller drives the stepper motor 501 to rotate in the same direction so that the clamping force is greater than the second preset threshold value, wherein the first preset threshold value is greater than the second preset threshold value. The stepper motor 501 can be a small servo stepper motor 501, the bidirectional lead screw 503 can be a trapezoidal lead screw, and the force sensor can be a miniature piezoresistive force sensor; the gripper 505 can be made of food-grade silicone with a Shore hardness between 30-40A, for example, 35A, and the bidirectional lead screw 503 can be made of stainless steel; The operating process is as follows: When a mango needs to be clamped, the controller controls the rotation of the stepper motor 501, which drives the two bases 502 via the bidirectional screw 503 to move, closing the clamping jaws 505. The force sensor monitors the clamping force in real time and provides feedback to the controller. The controller adjusts the rotation direction of the stepper motor 501 based on the feedback force value to maintain the clamping force between 0.5 and 2N. By adopting this technical solution, the present invention realizes adaptive clamping of mangoes of different maturity through the flexible clamping claws 505 and the force feedback system, avoiding damage to the fruit caused by excessive clamping force, while ensuring the stability of clamping and reducing the damage rate of mangoes during the picking process.

[0033] In another technical solution, the conveying and collecting mechanism includes: The inner wall of the conveying pipe is paved with a silicone pad with a thickness of ≥5mm, such as a 5mm or 6mm thick silicone pad, to cushion the impact of the fruit during its falling process; A collecting container 3 is placed on the support base 2. A weight sensor is provided on the support base 2. The weight sensor is connected to the controller and is used to monitor the total weight of the mangoes in the collecting container 3 in real time. a prompt module connected to the controller, wherein when the weight sensor detects that the weight is greater than or equal to a preset weight threshold, the controller sends a replacement signal to the prompt module, the prompt module including at least one of the following: an audible and visual alarm, and a wireless communication module, the wireless communication module being configured to send a replacement instruction to an external device; the prompt module may include an audible and visual alarm, and may also include a wireless communication module, the wireless communication module being configured to send a replacement instruction to an external device, the preset weight threshold being set according to the capacity of the collection container 3, for example, 20 kg or 25 kg; The silicone pad can be made of food-grade silicone, the weight sensor can be a strain gauge weighing sensor, the sound and light alarm can be a common combination of buzzer and indicator light, and the wireless communication module can be a Bluetooth module or a WiFi module; The process is as follows: After being picked, mangoes fall into collection container 3 through a conveyor pipe. A weight sensor monitors the weight of the mangoes in the container in real time and transmits the data to a controller. When the weight reaches a preset threshold, the controller sends a signal to a prompt module, which notifies the operator to replace collection container 3 through audible and visual alarms or wireless communication. This technical solution realizes the automatic collection of mangoes and the container replacement prompt function, improves the collection efficiency, and reduces the collision damage of the fruit during transportation, ensuring the quality of the fruit.

[0034] In another technical solution, it also includes: A linear displacement platform is connected to the lens assembly, the linear displacement platform is driven by a telescopic motor, and the telescopic motor is connected to the controller; A three-dimensional profile sensor is located on the side of the laser emission module and is used to scan the surface curvature of the mango fruit in real time and generate surface concave and convex feature data; The dynamic focusing algorithm module is built into the controller and calculates the optimal focusing distance of the detection area based on the surface concave and convex feature data, which satisfies: When the curvature radius of the fruit surface is less than 50 mm, shorten the detection distance to 5-8 cm (such as 6 cm or 7 cm) to increase the laser energy density; When the curvature radius is greater than 150 mm, the detection distance is extended to 12-15 cm to expand the coverage of ethylene molecules; The linear displacement platform drives the lens group to move along the optical axis according to the controller's instructions, so that the detection area matches the curvature radius of the fruit surface in real time, and the distance adjustment response time is less than 100ms; The linear displacement platform can be constructed using a linear guide slide, the telescopic motor can be a 501 stepper motor or a servo motor, and the 3D profile sensor can be a laser triangulation sensor or a structured light sensor. The linear displacement platform's guide rails can be made of aluminum alloy or stainless steel to ensure stable and precise movement. During assembly, the linear displacement platform is mounted on the bracket of the laser emission module and lens assembly, and the 3D profile sensor is installed to the side of the laser emission module to ensure accurate scanning of the fruit surface. The operating process is as follows: A 3D profile sensor scans the mango fruit surface in real time, generating surface curvature data and transmitting it to a controller. The controller's dynamic focusing algorithm calculates the optimal focusing distance based on this data. The controller then issues a command to the linear displacement stage, which drives the lens assembly along the optical axis to the appropriate position, focusing the laser beam on the optimal detection area. For example, if the surface curvature radius of the fruit is detected to be small, such as 40mm, the controller instructs the linear displacement stage to shorten the detection distance to 6cm, ensuring sufficient laser energy density for accurate ethylene content detection. By adopting this technical solution, through dynamic focusing algorithm and linear displacement platform, the detection area can match the curvature of the fruit surface in real time, improving the accuracy and stability of ethylene concentration detection, and ensuring that ethylene content can be accurately detected on mango surfaces of different shapes.

[0035] In another technical solution, the three-dimensional profile sensor includes: A dual-band laser scanning head, which includes a 940nm wavelength structured light projector and a 1550nm wavelength laser radar. The structured light projector is used to generate phase-coded stripes on the mango surface, and the laser radar is used to obtain millimeter-level depth point clouds on the fruit surface. Polarization filter set, which is set in front of the sensor receiving end, with its transmission axis direction orthogonal to the polarization direction of the scattered light in the orchard environment, to suppress leaf reflection and water mist interference; A dynamic calibration module, built into the sensor housing, adjusts the structured light projection power and lidar sampling frequency based on ambient light intensity to maintain a signal-to-noise ratio above 20dB. The controller fuses the phase data of the phase-coded stripes with the lidar point cloud data and reconstructs the fruit surface geometric model through a curvature compensation algorithm.

[0036] The dual-band laser scanning head can use the existing structured light and laser radar integrated module on the market, the polarization filter group can use a polarizer, and the dynamic calibration module can use a microprocessor and related circuits; The operating process is as follows: A dual-band laser scanner emits 940nm structured light and 1550nm laser light. The structured light projector generates phase-coded stripes on the mango surface, and the lidar acquires depth point cloud data of the fruit surface. A polarizing filter array filters out ambient scattered light to reduce interference. The dynamic calibration module adjusts the structured light projection power and lidar sampling frequency based on ambient light intensity. For example, when the light intensity is strong, the structured light projection power is increased to improve signal strength. The controller receives the phase data and point cloud data and uses a curvature compensation algorithm to reconstruct the fruit surface geometry model, thereby obtaining information such as the fruit surface curvature radius. By adopting this technical solution, accurate scanning of the geometric features of the mango fruit surface is achieved through a dual-band laser scanning head and a polarization filter set, providing accurate data support for dynamic focusing, improving the scanning accuracy and anti-interference ability, and ensuring that the fruit surface data can be accurately obtained under different environmental conditions.

[0037] In another technical solution, the dynamic calibration module includes: The light intensity grading unit (the light intensity grading unit can use a photoresistor or illuminance sensor to detect the ambient light intensity, and adjust the structured light projection power and the lidar sampling frequency through circuits such as operational amplifiers and relays) adjusts the structured light projection power and the lidar sampling frequency according to the ambient light intensity. Its configuration is as follows: When the ambient light intensity is less than 10,000 lux, the structured light projection power is controlled to 50 mW and the lidar sampling frequency is 10 Hz; When the illumination intensity is 10,000 lux or less than 80,000 lux, the structured light power is dynamically adjusted according to P = 50 + n × (I - 10,000), and the lidar sampling frequency is adjusted to 20 Hz, where I is the ambient light intensity, P is the structured light projection power, and n is 0.00286 mW / lux. When the light intensity is ≥80,000 lux, the structured light projection power is adjusted to 250 mW, the lidar anti-interference mode is activated and the sampling frequency is adjusted to 50 Hz; The operating process is as follows: The illuminance sensor detects ambient light intensity in real time and transmits the data to the light intensity classification unit. The light intensity classification unit automatically adjusts the structured light projection power and the lidar sampling frequency according to the different light intensity ranges. For example, when the light intensity is 20,000 lux, the structured light projection power P = 50 + 0.00286 × (20,000 - 10,000) = 78.6 mW, and the lidar sampling frequency is adjusted to 20 Hz. This technical solution enables the sensor to maintain a high signal-to-noise ratio under varying light intensities, ensuring the accuracy of scanned data. This technical feature adapts to varying ambient lighting conditions through graded light intensity adjustment, improving the operating stability and data accuracy of the 3D profile sensor in various lighting environments and ensuring precise scanning of fruit surfaces.

[0038] In another technical solution, the dynamic calibration module further includes a multi-parameter coupling compensation unit, and the configuration of the multi-parameter coupling compensation unit is: The segmented humidity compensation module divides the structured light power additional humidity compensation coefficient into three levels of compensation strategies according to the ambient humidity data: Dry mode, ambient humidity RH < 65%: humidity compensation coefficient k3 = 1.0; Normal mode, ambient humidity 65% ​​≤ RH ≤ 90%: k3 = 1 + 0.03 × (RH - 65); High humidity mode, ambient humidity is RH>90%: k3=1.75+0.12×(RH-90); The working process is as follows: The humidity sensor detects the ambient humidity in real time and transmits it to the segmented humidity compensation module. The module calculates the corresponding humidity compensation coefficient k3 based on the current humidity range and then adjusts the structured light power to the original power multiplied by k3. For example, when the ambient humidity is 70% RH, in normal mode, k3 = 1 + 0.03 × (70-65) = 1.15, and the structured light power is adjusted to 1.15 times the original power. This technical solution compensates for the effect of ambient humidity on structured light transmission, ensuring the sensor's proper operation in varying humidity environments. This segmented humidity compensation strategy reduces the impact of ambient humidity on structured light power, improving the 3D profile sensor's operational stability and scanning accuracy in varying humidity environments, ensuring accurate data collection on the fruit surface.

[0039] In another technical solution, the multi-parameter coupling compensation unit is further configured as follows: a temperature-power coupling module, which adds a temperature compensation coefficient k4=1-0.015×(T-25) to the structured light power according to the ambient temperature T. When T>25°C, the power is dynamically reduced to suppress temperature drift. For example, when the ambient temperature is 30°C, k4=1-0.015×(30-25)=0.925, and the structured light power is adjusted to 0.925 times the original power. The operating process is as follows: The temperature sensor detects the ambient temperature in real time and transmits it to the temperature-power coupling module. The module calculates the temperature compensation coefficient k4 based on the current temperature. When the temperature is greater than 25°C, the structured light power is adjusted to the original power multiplied by k4, achieving dynamic power reduction. For example, when the temperature is 35°C, k4 = 1-0.015×(35-25) = 0.85, and the structured light power is reduced to 85% of the original power to suppress the impact of temperature drift on structured light projection. This technical solution reduces the impact of ambient temperature changes on structured light power through temperature-power coupling compensation, improves the working stability of the three-dimensional profile sensor in different temperature environments, and ensures the accuracy and reliability of the fruit surface scanning data.

[0040] In another technical solution, the composite soundproof enclosure includes, from the inside to the outside: Sound absorbing layer, which is made of gradient density porous foam material, with density changing from 800 kg / m³ in the inner layer to 200 kg / m³ in the outer layer, and thickness of 10-15 mm; Damping layer, which is composed of butyl rubber and carbon fiber woven mesh, with a thickness of 2-3 mm; The reflective layer, an aluminum alloy plate with a micro-prism array on its surface, can reflect sound waves and further enhance the sound insulation effect; Directional microphone array, which consists of four MEMS microphones arranged in a circular pattern at equal intervals in the center of the soundproof enclosure. Its directivity meets the following requirements: main lobe width ≤ 30°, focused on the center of the detection area, and sidelobe suppression ratio > 20 dB to reduce lateral noise interference; An adaptive noise cancellation module, embedded in the signal processing unit, extracts the ambient noise reference signal in real time through an LMS filtering algorithm and cancels the common-mode noise component from the microphone array output signal; The sound-absorbing layer, damping layer, and reflective layer are sequentially attached to the inner wall of the soundproof enclosure, and the directional microphone array is fixed in the center of the soundproof enclosure to ensure that it points to the detection area. The operating process is as follows: When ambient noise enters the soundproof enclosure, the sound-absorbing layer absorbs most of the noise, the damping layer suppresses the noise generated by the enclosure's vibration, and the reflective layer reflects the remaining sound waves. A directional microphone array focuses on the detection area and receives the photoacoustic signal. The adaptive noise cancellation module uses the LMS filtering algorithm to offset the ambient noise from the microphone array's output signal, producing a clear photoacoustic signal. By adopting this technical solution, the composite soundproof cover and directional microphone array can effectively suppress the interference of environmental noise, improve the reception quality of photoacoustic signals, make the ethylene concentration detection more accurate, and ensure the reliability of maturity judgment.

[0041] The number of devices and processing scales described here are used to simplify the description of the present invention. Applications, modifications and variations of the intelligent mango harvesting device of the present invention will be obvious to those skilled in the art.

[0042] Although the embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the description and implementation methods. They can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and illustrations shown and described herein.

Claims

1. An intelligent mango harvesting device, comprising a traveling mechanism, a grabbing mechanism disposed on the traveling mechanism, and a conveying and collecting mechanism disposed below the grabbing mechanism, characterized in that: The device also includes an ethylene content detection module, which includes a non-contact ethylene sensor. The non-contact ethylene sensor includes: A laser emission module, which is used to emit a pulsed laser beam to the detection area. The emission wavelength of the laser emission module is 10.3-10.6 μm; A lens assembly is provided on the outgoing optical path of the laser emission module. The lens assembly is used to focus the pulsed laser beam into a preset spot size and irradiate the air 5-15 cm from the surface of the mango fruit to form an ethylene molecule detection area. In the detection area, the pulsed laser beam interacts with the ethylene molecules released by the mango to generate a photoacoustic signal. A photoacoustic signal receiving unit, comprising a soundproof enclosure and a microphone, wherein the soundproof enclosure is used to suppress environmental noise interference, and the microphone is used to receive the photoacoustic signal generated in the detection area; a signal processing unit electrically connected to the microphone and configured to process the received photoacoustic signal to generate a digital electrical signal; The controller is connected to the signal processing unit through a data bus. The controller receives the digital electrical signal and converts the digital electrical signal into an ethylene gas concentration value, and compares the ethylene gas concentration value with a preset ethylene concentration threshold to determine the maturity of the mango.

2. The intelligent mango harvesting device according to claim 1, characterized in that: It also includes an environmental monitoring module, which is connected to the controller. The environmental monitoring module is equipped with a temperature and humidity sensor. The controller uses a temperature and humidity compensation algorithm to correct the impact of the environment on ethylene concentration detection. The specific formula is: C 修正 = C 实测 ×(1+ k 1D T + k 2D RH ) in, k 1. k 2 are the temperature compensation coefficient and humidity compensation coefficient, respectively, obtained through laboratory calibration, Δ RH is the difference between the current humidity and the standard humidity, Δ T is the difference between the current temperature and the standard temperature, C 实测 is the original detection value of ethylene concentration directly obtained by the photoacoustic signal receiving unit, C 修正 is the final concentration value after adjustment based on the temperature and humidity compensation algorithm.

3. The intelligent mango harvesting device according to claim 1, characterized in that: The gripping mechanism includes: Two clamping jaws, which are made of silicone material, and the inner wall of the clamping jaws is provided with a wave-shaped buffer groove with a depth of 0.3-0.5cm and a spacing of 1cm; A drive assembly includes a stepper motor connected to a controller, a bidirectional screw connected to the stepper motor, and two bases provided on the bidirectional screw. The roots of the two clamping jaws are respectively connected to the two bases so that the clamping force between the two clamping jaws can be adjusted by rotating the bidirectional screw. A force sensor is embedded in the root of the clamping jaw and connected to the controller. The force sensor is used to monitor the clamping force in real time and feed it back to the controller. When the clamping force monitored by the force sensor is ≥ a first preset threshold, the controller drives the stepper motor to open in the opposite direction so that the clamping force is less than the first preset threshold. When the clamping force monitored by the force sensor is ≤ a second preset threshold, the controller drives the stepper motor to rotate in the same direction so that the clamping force is greater than the second preset threshold, wherein the first preset threshold is greater than the second preset threshold.

4. The intelligent mango harvesting device according to claim 3, characterized in that: The conveying and collecting mechanisms include: The inner wall of the delivery pipeline shall be paved with a silicone pad with a thickness of ≥5mm; A collection container is placed on a support base, wherein a weight sensor is provided on the support base, the weight sensor is connected to the controller, and the weight sensor is used to monitor the total weight of the mangoes in the collection container in real time; The prompt module is connected to the controller. When the weight sensor detects that the weight is ≥ the preset weight threshold, the controller sends a replacement signal to the prompt module. The prompt module includes at least one of the following: an audible and visual alarm, and a wireless communication module. The wireless communication module is used to send a replacement instruction to an external device.

5. The intelligent mango harvesting device according to claim 1, characterized in that: Also includes: A linear displacement platform is connected to the lens assembly, the linear displacement platform is driven by a telescopic motor, and the telescopic motor is connected to the controller; A three-dimensional profile sensor is located on the side of the laser emission module and is used to scan the surface curvature of the mango fruit in real time and generate surface concave and convex feature data; The dynamic focusing algorithm module is built into the controller and calculates the optimal focusing distance of the detection area based on the surface concave and convex feature data, which satisfies: When the curvature radius of the fruit surface is less than 50 mm, shorten the detection distance to 5-8 cm to enhance the laser energy density; When the curvature radius is greater than 150 mm, the detection distance is extended to 12-15 cm to expand the coverage of ethylene molecules; The linear displacement platform drives the lens group to move along the optical axis according to the controller instructions, so that the detection area matches the curvature radius of the fruit surface in real time, and the distance adjustment response time is less than 100ms.

6. The intelligent mango harvesting device according to claim 5, characterized in that: 3D profile sensors include: A dual-band laser scanning head, which includes a 940nm wavelength structured light projector and a 1550nm wavelength laser radar. The structured light projector is used to generate phase-coded stripes on the mango surface, and the laser radar is used to obtain millimeter-level depth point clouds on the fruit surface. Polarization filter set, which is set in front of the sensor receiving end, with its transmission axis direction orthogonal to the polarization direction of the scattered light in the orchard environment, to suppress leaf reflection and water mist interference; A dynamic calibration module, built into the sensor housing, adjusts the structured light projection power and lidar sampling frequency based on ambient light intensity to maintain a signal-to-noise ratio above 20dB. The controller fuses the phase data of the phase-coded stripes with the lidar point cloud data and reconstructs the fruit surface geometric model through a curvature compensation algorithm.

7. The intelligent mango harvesting device according to claim 6, characterized in that: The dynamic calibration module includes: The light intensity grading unit is configured as follows: When the ambient light intensity is less than 10,000 lux, the structured light projection power is controlled to 50 mW and the lidar sampling frequency is 10 Hz; When the illumination intensity is 10,000 lux or less than 80,000 lux, the structured light power is dynamically adjusted according to P = 50 + n × (I - 10,000), and the lidar sampling frequency is adjusted to 20 Hz, where I is the ambient light intensity, P is the structured light projection power, and n is 0.00286 mW / lux. When the light intensity is ≥80,000 lux, the structured light projection power is adjusted to 250 mW, the lidar anti-interference mode is activated and the sampling frequency is adjusted to 50 Hz.

8. The intelligent mango harvesting device according to claim 7, characterized in that: The dynamic calibration module further includes a multi-parameter coupling compensation unit, and the configuration of the multi-parameter coupling compensation unit is as follows: The segmented humidity compensation module divides the structured light power additional humidity compensation coefficient into three levels of compensation strategies according to the ambient humidity data: Dry mode, ambient humidity RH < 65%: humidity compensation coefficient k3 = 1.0; Normal mode, ambient humidity 65% ​​≤ RH ≤ 90%: k3 = 1 + 0.03 × (RH - 65); High humidity mode, ambient humidity is RH>90%: k3=1.75+0.12×(RH-90).

9. The intelligent mango harvesting device according to claim 8, characterized in that: The multi-parameter coupling compensation unit is further configured as follows: a temperature-power coupling module, which adds a temperature compensation coefficient k4=1-0.015×(T-25) to the structured light power according to the ambient temperature T, and dynamically reduces the power when T>25℃ to suppress temperature drift.

10. The intelligent mango harvesting device according to claim 1, characterized in that: The composite soundproof cover includes: Sound absorbing layer, which is made of gradient density porous foam material, with density changing from 800 kg / m³ in the inner layer to 200 kg / m³ in the outer layer, and thickness of 10-15 mm; Damping layer, which is composed of butyl rubber and carbon fiber woven mesh, with a thickness of 2-3 mm; A reflective layer, comprising an aluminum alloy plate with a microprism array on its surface; Directional microphone array, which consists of four MEMS microphones arranged in a circular pattern at equal intervals in the center of the soundproof enclosure. Its directivity meets the following requirements: main lobe width ≤ 30°, focused on the center of the detection area, and sidelobe suppression ratio > 20 dB to reduce lateral noise interference; The adaptive noise cancellation module is embedded in the signal processing unit and uses the LMS filtering algorithm to extract the ambient noise reference signal in real time and cancel the common-mode noise component from the microphone array output signal.

Images