Shading Compensation Method and System for Multi-Angle Segmented UVC Irradiation
By acquiring three-dimensional data of crop canopy, adjusting the power and pose of UVC lamp arrays in segments, constructing a light field contribution matrix, and optimizing the power vector of lamp segments, the problem of light blind spots caused by leaf shading in high-canopy crops was solved, achieving dose uniformity and safety, and improving disease control effects and equipment adaptability.
Patent Information
- Application Number
- CN202611501088.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-24
- Publication Date
- 2026-11-17
AI Technical Summary
Existing UVC plant protection equipment has light blind spots caused by leaf shading in tall canopy crops, resulting in insufficient or excessive dosage in some areas. Current technologies can only compensate for this by increasing the total power, which is not effective enough to achieve uniform coverage and poses a risk of crop damage.
By acquiring three-dimensional point cloud data of crop canopy, the power and pose of UVC lamp arrays are adjusted in segments to construct a light field contribution matrix, optimize the lamp segment power vector, and combine visibility and dose thresholds to identify and compensate for under-covered areas, thereby achieving multi-angle directional irradiation.
It achieves dose uniformity and safety control in complex canopies, avoids excessive irradiation and blind spots, improves disease control and crop safety, and reduces energy consumption and operating costs.
Smart Images

Figure CN123124184A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of optical plant protection technology in facility agriculture, specifically to a shading compensation method and system for multi-angle segmented UVC irradiation. Background Technology
[0002] Using UVC short-wave ultraviolet radiation to inhibit or kill pathogenic spores on crop leaves is an effective technical means to replace chemical pesticides in greenhouse agriculture. As greenhouse cultivation develops towards tall, elevated crops, the crop canopy has transformed from a low, flat structure to a complex three-dimensional structure with multiple layers and leaf orientations. The phenomenon of mutual shading between leaves, between leaves and fruits, or between petioles is very common, which places higher demands on the accuracy of UVC irradiation dosage and the area coverage.
[0003] Existing UVC plant protection equipment is mainly divided into two categories: fixed lamp stands and mobile robots. Most of them adopt an integrated long lamp tube structure, and control the irradiation dose by uniformly adjusting the output power. Some improvements use measuring points arranged in a two-dimensional plane to detect irradiance and correct the uniformity of the overall light intensity. With the development of technology, segmented dimming schemes have also emerged, which divide the lamp array into several independent control units, and can roughly match the irradiation needs of different canopy heights in the height direction.
[0004] However, the diverse spatial orientations of leaves in tall-canopy crops mean that upper leaves, petioles, and fruits create continuous geometric shading on the undersides of middle and lower leaves, resulting in numerous blind spots where light cannot directly reach even under a single illumination position. Current technologies address this shading by increasing the overall power of the lighting system, relying on diffused and scattered light to supplement the dose in these blind spots. However, the increased power causes the front and outer leaves of the canopy to receive excessive UVC doses far exceeding the control requirements. If this exceeds the crop's tolerance threshold, it can lead to plant damage. Furthermore, simply increasing power has limited effect on improving the visibility blind spots caused by geometric shading; the undersides of leaves and inner canopies still fail to reach the minimum effective control dose, ultimately resulting in wasted light on the front and insufficient dose in the blind spots. Summary of the Invention
[0005] A method for shading compensation for multi-angle segmented UVC irradiation includes the following steps:
[0006] S1. Acquire the three-dimensional point cloud data of the target crop canopy, and perform surface segmentation on the three-dimensional point cloud data to obtain several target surface units. Assign a spatial location to each target surface unit. Surface normal Visibility and prescription dosage Prescription dosage Determined based on the photorepair spectrum of the target pathogen;
[0007] S2. Obtain the current illumination pose of the segmented UVC lamp array. Combining the geometric parameters of the segmented UVC lamp array with the spatial position of each target surface unit Surface normal Visibility Establish the current illumination pose Light field contribution matrix under , Indicates the first UVC lamp segment in illumination position Next to the Target surface unit The effective irradiance contribution value, ;
[0008] S3, Target irradiance of each target surface unit To optimize the objective, based on the light field contribution matrix A constrained optimization model with regularization terms is constructed to obtain the optimal power vector for each UVC lamp segment. ,in This represents the total number of segments in the UVC lamp array;
[0009] S4. Calculate the predicted cumulative dose for each target surface unit based on the optimal power vector and the light field contribution matrix. This will predict the cumulative dose. With prescription dosage Visibility The model is compared with a preset threshold to identify under-covered target surface units. If an under-covered target surface unit exists, it is selected from a preset candidate pose set. Second irradiation pose in the middle screening Based on the second illumination pose The corresponding optical field contribution matrix only performs directional dose compensation for under-covered target surface units.
[0010] Preferably, the total number of segments in the UVC lamp array Determined according to the following formula:
[0011]
[0012] in, The effective height of the target canopy, To achieve the preset spatial resolution, each segment of the UVC lamp array is equipped with an independent constant current or constant power drive unit, optical partitions are set between adjacent lamp segments, and reflective cavities and light-shielding skirts are set on the outside of the lamp array.
[0013] Preferably, based on the robot pose Based on the geometric model of the lamp segments, the illumination distance from each lamp segment to each target surface unit is calculated. With the angle of incidence Combined with visibility The effective irradiance contribution value was calculated. .
[0014] Preferred light field contribution matrix The distance is obtained through pre-calibration, and the median value of the target working distance is selected as the standard distance. At standard distance The system uses multiple measuring points arranged at different distances above and below the lamps to illuminate the lamp segments one by one and measure the irradiance, generating calibration data at different distances. The calibration environment records include temperature, humidity, lamp preheating time, and power supply status.
[0015] Preferably, the objective function of the constrained optimization model is:
[0016]
[0017] in, For the first The weighting coefficients of each target surface unit. This is the power vector of the lamp segment at the previous moment. These are the coefficients for the power smoothing regularization term. This is the coefficient for the total power penalty term.
[0018] Preferably, the constraints of the constrained optimization model include:
[0019]
[0020] In the formula, For the first The maximum rated power of the segment lamp, For the first Operating temperature of segment lamps This is the upper limit of temperature. To predict leakage irradiance, For safe leakage threshold, For the first The maximum tolerated dose for each target surface unit.
[0021] Preferably, the light field reconstruction effect is quantitatively evaluated using two indicators: uniformity and coverage.
[0022] Uniformity is defined as ;in, The 10th percentile value of the cumulative dose for all target surface units. The average cumulative dose for all target surface units; uniformity can also be selected based on project acceptance requirements, using the ratio of minimum dose to average dose. Perform calculations;
[0023] Coverage is defined as: ;in, The number of surface units required to accumulate the dose to reach the target dose. The total number of target surface units. This is the dosage target achievement factor.
[0024] In the initial stage of the project, a uniformity target can be set as follows: The coverage target is The final threshold is determined by the experimental results of crop varieties and application scenarios.
[0025] Preferably, the determination criterion for under-covered target surface units is:
[0026]
[0027] in, To preset the visibility threshold, The dose compliance coefficient is used; target surface units that meet any of the conditions are judged as under-covered target surface units.
[0028] Preferably, the second irradiation pose is selected. The rules are:
[0029] In the candidate pose set In the middle, select the pose with the largest reduction in corresponding dose deficit and the smallest increase in movement distance;
[0030] Second illumination pose When dose compensation is performed, the cumulative dose to each target surface unit shall not exceed its maximum tolerated dose. .
[0031] Preferably, the compensation task generation of the second irradiation pose is associated with the crop target unit dose status. Based on the cumulative UVC dose of each target unit, the crop tolerance margin and the degree of dark period failure are introduced as constraints, and the cumulative dose after compensation does not exceed the upper limit of the crop tolerance dose.
[0032] Preferably, the canopy distance and attitude data are updated every 100-500ms during the irradiation process, and the lamp power adjustment adopts a ramp limit; if a change in pose is detected that causes a mismatch in the light field model, the output power of the lamp segment is reduced first, and then the optimal power vector is resolved.
[0033] An occlusion compensation system for multi-angle segmented UVC illumination includes a 3D perception module, a segmented UVC lamp array module, a pose execution module, a light field calibration module, a constraint optimization solution module, a pose planning module, and a main control module.
[0034] The 3D perception module is used to acquire 3D point cloud data of the target crop canopy, perform surface segmentation, and output the parameters of each target surface unit to the main control module;
[0035] The segmented UVC lamp array module is divided into N independently controlled UVC lamp segments along the height and / or length direction. Each lamp segment is equipped with an independent drive unit to receive power commands from the main control module and output the corresponding irradiance.
[0036] The pose execution module is used to drive the segmented UVC lamp array module to adjust the irradiation pose and feed the real-time pose data back to the main control module.
[0037] The light field calibration module is used to calibrate the light field contribution matrix of each lamp segment to each target surface unit under different illumination poses, and to store the calibration data to the main control module.
[0038] The constraint optimization solution module receives the target surface element parameters and the optical field contribution matrix, constructs the constraint optimization model, solves the optimal power vector, and outputs it to the main control module.
[0039] The pose planning module is used to identify undercovered targets based on the predicted dose under optimal power, and to plan a second compensation pose from the candidate pose set and output it to the main control module.
[0040] The main control module is used to schedule each module to execute the shading compensation method for multi-angle segmented UVC irradiation.
[0041] The beneficial effects of this invention are as follows:
[0042] 1. This invention enables dose control for complex crop canopies, adapts to the diverse leaf postures and layered structures of three-dimensional canopies, matches the irradiation requirements of different canopy heights through segmented independent light adjustment, and eliminates irradiation blind spots caused by leaf shading by combining multi-position directional supplementary irradiation. This ensures that the hidden areas reach the effective control dose while avoiding excessive irradiation of the front and outer areas of the canopy, thus improving the disease control effect and crop growth safety.
[0043] 2. It abandons the model of relying on increasing total power to compensate for shading defects, and achieves the coverage target by coordinating power optimization allocation and pose adjustment. This can effectively reduce operating energy consumption and light source output load, reduce device thermal decay and aging speed, and reduce long-term operating costs in facility plant protection scenarios.
[0044] 3. Through calibration logic and closed-loop control process, it can adapt to the canopy morphology of different crop varieties and different growth stages. It can autonomously complete the dynamic reconstruction of light field and shading compensation without repeated manual debugging, thus improving the scene adaptability of the equipment. Attached Figure Description
[0045] Figure 1This is a block diagram illustrating the principle of calculating the light field dose of a three-dimensional canopy target using a segmented light array according to the present invention.
[0046] Figure 2 Flowchart for solving the power and evaluating the effect of segmented light array;
[0047] Figure 3 The flowchart shows the optimization process for second illumination pose compensation under single pose under-coverage. Detailed Implementation
[0048] The application principle of the present invention will be further described below with reference to the accompanying drawings and specific embodiments.
[0049] Example 1:
[0050] like Figure 1-3 As shown, the occlusion compensation system for multi-angle segmented UVC illumination includes a three-dimensional perception module, a segmented UVC lamp array module, a pose execution module, a light field calibration module, a constraint optimization solution module, a pose planning module, and a main control module.
[0051] In this embodiment, the three-dimensional perception module uses a depth camera or lidar to collect three-dimensional point cloud data of the target crop canopy, and completes point cloud preprocessing, surface segmentation and unit parameter calculation.
[0052] The segmented UVC lamp array module is divided into N independently controlled UVC lamp segments along the height and / or length direction. Each segment is equipped with an independent constant current drive unit and dimming circuit. Optical partitions are set between adjacent lamp segments to suppress inter-segment light crosstalk. Reflective cavities and light-shielding skirts are set on the outside of the lamp array to constrain the light output range and improve light utilization.
[0053] The pose execution module adopts a mobile robot mounting mechanism, which can drive the segmented UVC lamp array to complete translation and rotation, adjust the illumination position and angle, and provide real-time feedback of pose data;
[0054] The light field calibration module includes a metrologically calibrated UVC radiometer and calibration station, which is used to measure the irradiance distribution of each lamp segment under different poses and distances, and generate and store the light field contribution matrix reference data.
[0055] The constrained optimization solution module is used to construct an optimization model with multiple constraints and solve for the optimal power output value of each lamp segment.
[0056] The pose planning module is used to identify under-covered target areas and plan a second illumination pose for compensation.
[0057] Example 2:
[0058] The occlusion compensation method based on the above system includes the following steps:
[0059] S1. Canopy 3D Modeling and Target Surface Unit Division: The 3D perception module scans the target crop canopy to acquire raw 3D point cloud data. After denoising, filtering, and registration preprocessing, the canopy surface is divided into meshes to obtain several target surface units. ,in , This represents the total number of surface units.
[0060] Then, corresponding parameters are assigned to each target surface unit: spatial position. The three-dimensional spatial coordinates of the element center point; surface normal. The unit normal vector, calculated by fitting a plane to the unit neighborhood point cloud, is used to characterize the leaf orientation; visibility. Based on the current irradiation direction, the proportion of the unit surface area that can be directly irradiated by light, calculated using the ray projection method, has a value range of [0,1]; prescription dose. The biologically equivalent effective dose is determined based on the photorepair spectrum of the target pathogen and the UVC inactivation spectrum and photorepair risk curve of the target pathogen. Its value takes into account the lethal dose threshold of the pathogen and the photoreactivation repair window, and is determined based on achieving the target biological control effect, rather than simply the physical irradiation integral value.
[0061] Among them, the photoremediation spectrum weight is a dose equivalence coefficient determined based on the biological effects of the target pathogen. The determination basis includes pathogen type, wavelength response, and field calibration. The pathogen type is selected according to the type of target pathogen. The wavelength response is determined based on the response difference between the UVC inactivation spectrum and the photoreactivation spectrum of the target pathogen, and the weight coefficient of each characteristic wavelength is determined. The field calibration is performed by correcting the standard spectrum weight through field biological testing or irradiation environment correction to form the final weight applicable to the field.
[0062] S2. Construction of the light field contribution matrix under the current pose: The pose execution module acquires the current illumination pose of the segmented UVC lamp array. , pose This includes the spatial coordinates of the light array and its pitch and yaw attitude angles. The main control module calls the geometric parameters of the segmented UVC light array, combined with the spatial position of each target surface unit. Surface normal Calculate the first Segment light up to the first Irradiation distance of each target surface unit Angle of incidence of light ;in , This represents the total number of segments in the UVC lamp array.
[0063] Combined with the visibility of the target surface unit Construct the current illumination pose Light field contribution matrix under ,in Indicates the first Under unit power, the UVC lamp segment of the first segment... The effective irradiance contribution value generated by each target surface unit.
[0064] Light field contribution matrix The standard distance is obtained by combining pre-calibration and online correction, with the median value of the target working distance pre-selected as the standard distance. At standard distance Measurement points are arranged in an array within multiple distance planes above and below the lamps. Each UVC lamp segment is illuminated sequentially, and the irradiance value at each measurement point is measured using a metrologically calibrated UVC radiometer. The temperature, humidity, lamp preheating time, and power supply status of the calibration environment are recorded to form benchmark calibration data. During actual operation, interpolation corrections are performed based on the real-time irradiation distance and incident angle, combined with the benchmark calibration data, to obtain real-time... .
[0065] In this embodiment, the light field contribution matrix is calibrated using the standard distance calibration method, where the standard distance is... Choose the median value of the product's target working distance, rather than an arbitrary fixed value; the standard distance for short canopy scenarios can be 0.30m, and the standard distance for high canopy side-illuminated scenarios can be 0.40m.
[0066] At least 5×3 measuring points should be arranged in each distance plane, or the spacing between measuring points should not exceed 100mm; calibration should be performed using a UVC radiometer that has been metrologically calibrated and whose spectral response covers the target wavelength band. During the calibration process, ambient temperature, ambient humidity, lamp preheating time, and power supply status should be recorded to ensure consistency of calibration conditions.
[0067] A simplified recalibration trigger mechanism is set up, which triggers simplified recalibration when any of the following conditions are met: the cumulative illumination time reaches a preset cycle, the irradiance deviation of the reference point exceeds 5%, the light source or UVC-resistant window is replaced, the window is cleaned, or the residual of the light field model continuously exceeds the limit. Simplified calibration points are preferentially selected from the maximum light field value point, edge value point, and representative intermediate point.
[0068] S3. Constraint optimization solution for segmented lamp array power: based on the target irradiance corresponding to each target surface unit. To optimize the objective, based on the light field contribution matrix A constrained optimization model with regularization terms is constructed, and the optimal power vector for each UVC lamp segment is obtained by solving the model. .
[0069] The objective function of the constrained optimization model is:
[0070]
[0071] in, For the first The weighting coefficients of each target surface unit are set with higher weighting values for difficult-to-illuminate or high-risk areas such as the back of the blade and the inner canopy. For the first Calculated value of actual irradiance received by each target surface unit; For the first The target irradiance of a target surface unit is determined by the prescription dose. It is calculated from the irradiation duration; This is the power vector of the lamp segment at the previous moment; The power smoothing regularization coefficient is used to constrain the instantaneous jump amplitude of power and protect the drive circuit and light source. This is the total power penalty term coefficient, used to reduce the total output power while meeting dosage requirements.
[0072] The constraints of the constrained optimization model include:
[0073]
[0074] in, For the first The maximum rated power of the segment lamp; The operating temperature of the k-th lamp is... This refers to the maximum permissible operating temperature of the luminaire. The predicted leakage irradiance is located outside the light array. The human safety leakage threshold; For the predicted cumulative dose of the j-th target surface unit, This represents the maximum tolerable dose for the crop in the corresponding unit. The optimal power vector under the current pose is obtained by solving the constrained optimization model using a quadratic programming algorithm. .
[0075] The light field reconstruction effect is quantitatively evaluated using two indicators: uniformity and coverage.
[0076] Uniformity is defined as ;in, The 10th percentile value of the cumulative dose for all target surface units. The average cumulative dose for all target surface units; uniformity can also be selected based on project acceptance requirements, using the ratio of minimum dose to average dose. Perform calculations;
[0077] Coverage is defined as: ;in, The number of surface units required to accumulate the dose to reach the target dose. The total number of target surface units. This is the dosage target achievement factor.
[0078] In the initial stage of the project, a uniformity target can be set as follows: The coverage target is The final threshold is determined by the experimental results of crop varieties and application scenarios.
[0079] S4. Occlusion Detection and Multi-Pose Orientation Compensation: Based on the optimal power vector With light field contribution matrix Calculate the predicted cumulative dose for each target surface unit. The cumulative dose will be predicted. With prescription dosage Compare and combine with visibility. Identify under-covered target surface units.
[0080] The criteria for determining under-covered target surface elements are:
[0081]
[0082] in, To preset the visibility threshold, The dose compliance coefficient is used; target surface units that meet any of the conditions are judged as under-covered target surface units.
[0083] If there is no under-covered target, the current pose and power are maintained; if there is an under-covered target, the pose planning module selects from a preset set of candidate poses. Second irradiation pose in the middle screening The selection rule is as follows: calculate the dose gap reduction and the increase in travel distance for each candidate pose, and select the pose with the largest dose gap reduction and the smallest increase in travel distance as the second irradiation pose. .
[0084] The pose execution module moves the segmented UVC lamp array to the second irradiation pose. The optical field contribution matrix under the corresponding pose is constructed, and the compensation power vector is solved again through constraint optimization. Directional dose compensation is performed only for under-covered target surface units, and the cumulative dose of all target surface units during the compensation process does not exceed their maximum tolerable dose. .
[0085] When identifying under-covered target surface units, the dose status data of the corresponding crop target units is called. Based on the cumulative effective dose, the actual remaining effective dose is obtained by combining the dark period failure coefficient and then the dose gap is calculated. When the compensation task is generated, the crop tolerance margin is used as the upper limit constraint to ensure that the cumulative dose of each target unit after compensation does not exceed the upper limit of the crop tolerance dose. After the compensation is completed, the newly added compensation dose is updated and written into the dose status of the corresponding crop target unit.
[0086] During illumination, the system updates the canopy distance and attitude data every 300ms. The lamp power adjustment uses a ramp limit to avoid instantaneous power jumps. If a change in pose is detected that causes the residual of the light field model to exceed a threshold, the output power of the lamp segment is reduced first, and then the optimal power vector is resolved. When the cumulative illumination time reaches a preset period, the reference irradiance deviation exceeds 5%, or the light source or window is replaced or cleaned, the recalibration process is triggered.
[0087] Example 3:
[0088] This embodiment is a specific application example of cucumber under high canopy side lighting.
[0089] The system employs a segmented UVC lamp array with an effective length of 1.8m, divided into 6 independently controlled lamp segments along the height direction. Each lamp segment is equipped with an independent constant current drive unit, and optical partitions are set between adjacent lamp segments. The outside of the lamp array is equipped with a reflective cavity and a light-shielding structure.
[0090] A 3D depth camera was used to scan the cucumber canopy to obtain a 3D point cloud of the canopy. The canopy surface was then divided into four target surface units: the upper front, the middle front, the lower front, and the underside of the leaves.
[0091] Select standard distance At the calibration station, each lamp segment is lit up one by one, and irradiance is measured at three distance planes of 0.30m, 0.40m and 0.50m respectively. The reference contribution matrix of each lamp segment is established, and the distance correction calibration is completed.
[0092] Under the first illumination pose, the system uses a constrained optimization model to solve for the optimal power of each lamp segment based on the canopy target unit parameters and the light field contribution matrix. After the solution is completed, the coverage effect is judged. The results show that the dose of the upper and middle front target units is up to standard, but the visibility of some target units on the back of the leaves is less than 0.5, which is due to under-coverage areas caused by geometric shading.
[0093] The system does not increase the output power of the front light segment, but controls the light array to move laterally by 0.22m within the range allowed by the walking channel, while adjusting the yaw angle of the light array so that the light array faces the back of the leaf; under the second irradiation pose, the corresponding light field contribution matrix is generated, and the power vector is solved again. Only the under-covered units on the back of the leaf are compensated for the directional dose. During the compensation process, the cumulative dose of all units does not exceed the UVC tolerance limit of the cucumber leaf.
[0094] All values in this embodiment are non-limiting engineering examples. The formal dose threshold, visibility threshold, and pose adjustment parameters will be ultimately determined by prototype testing and crop tolerance testing.
Claims
1. A shadowing compensation method for multi-angle segmented UVC irradiation, characterized by: Includes the following steps: S1, obtaining three-dimensional point cloud data of a target crop canopy, performing surface segmentation on the three-dimensional point cloud data to obtain a plurality of target surface units , assigning a spatial position to each target surface unit , a surface normal , a visibility rate , and a prescription dose ; the prescription dose is determined based on a light repair spectrum of a target pathogen S2. Obtain the current illumination pose of the segmented UVC lamp array. Combining the geometric parameters of the segmented UVC lamp array with the spatial position of each target surface unit Surface normal Visibility Establish the current illumination pose Light field contribution matrix under , Indicates the first UVC lamp segment in illumination position Next to the Target surface unit The effective irradiance contribution value; S3, Target irradiance of each target surface unit To optimize the objective, based on the light field contribution matrix A constrained optimization model with regularization terms is constructed to obtain the optimal power vector for each UVC lamp segment. ,in This represents the total number of segments in the UVC lamp array; S4. Calculate the predicted cumulative dose for each target surface unit based on the optimal power vector and the light field contribution matrix. This will predict the cumulative dose. With prescription dosage Visibility The model is compared with a preset threshold to identify under-covered target surface units. If an under-covered target surface unit exists, it is selected from a preset candidate pose set. Second irradiation pose in the middle screening Based on the second illumination pose The corresponding optical field contribution matrix only performs directional dose compensation for under-covered target surface units.
2. The shading compensation method for multi-angle segmented UVC irradiation according to claim 1, characterized in that: The total number of segments of the UVC lamp array Determined according to the following formula: in, The effective height of the target canopy, To achieve the preset spatial resolution, each segment of the UVC lamp array is equipped with an independent constant current or constant power drive unit, optical partitions are set between adjacent lamp segments, and reflective cavities and light-shielding skirts are set on the outside of the lamp array.
3. The shading compensation method for multi-angle segmented UVC irradiation according to claim 1, characterized in that: In step S2, based on the robot pose Based on the geometric model of the lamp segments, the illumination distance from each lamp segment to each target surface unit is calculated. With the angle of incidence Combined with the visibility rate The effective irradiance contribution value was calculated. .
4. The shading compensation method for multi-angle segmented UVC irradiation according to claim 1, characterized in that: The light field contribution matrix The distance is obtained through pre-calibration, and the median value of the target working distance is selected as the standard distance. At standard distance The system uses multiple measuring points arranged at different distances above and below the lamps to illuminate the lamp segments one by one and measure the irradiance, generating calibration data at different distances. The calibration environment records include temperature, humidity, lamp preheating time, and power supply status.
5. The shading compensation method for multi-angle segmented UVC irradiation according to claim 1, characterized in that: In S3, the objective function of the constrained optimization model is: in, For the first The weighting coefficients of each target surface unit. This is the power vector of the lamp segment at the previous moment. These are the coefficients for the power smoothing regularization term. This is the coefficient for the total power penalty term.
6. The shading compensation method for multi-angle segmented UVC irradiation according to claim 1, characterized in that: The constraints of the constrained optimization model include: In the formula, For the first The maximum rated power of the segment lamp, For the first Operating temperature of segment lamps This is the upper limit of temperature. To predict leakage irradiance, For safe leakage threshold, For the first The maximum tolerated dose for each target surface unit.
7. The shading compensation method for multi-angle segmented UVC irradiation according to claim 1, characterized in that: In S4, the determination condition for the under-covered target surface unit is: in, To preset the visibility threshold, The dose compliance coefficient is used; target surface units that meet any of the conditions are judged as under-covered target surface units.
8. The shading compensation method for multi-angle segmented UVC irradiation according to claim 1, characterized in that: In step S4, the second irradiation pose is selected. The rules are: In the candidate pose set In the middle, select the pose with the largest reduction in corresponding dose deficit and the smallest increase in movement distance; Second illumination pose When dose compensation is performed, the cumulative dose to each target surface unit shall not exceed its maximum tolerated dose. .
9. The shading compensation method for multi-angle segmented UVC irradiation according to claim 1, characterized in that: During the illumination process, the canopy distance and attitude data are updated every 100-500ms. The lamp power adjustment adopts a ramp limit. If a change in pose is detected that causes a mismatch in the light field model, the output power of the lamp segment is reduced first, and then the optimal power vector is resolved.
10. A shading compensation system for multi-angle segmented UVC irradiation, characterized in that: It includes a 3D perception module, a segmented UVC light array module, a pose execution module, a light field calibration module, a constraint optimization solution module, a pose planning module, and a main control module; The three-dimensional perception module is used to acquire three-dimensional point cloud data of the target crop canopy, perform surface segmentation, and output the parameters of each target surface unit to the main control module; The segmented UVC lamp array module is divided into N independently controlled UVC lamp segments along the height and / or length direction. Each lamp segment is equipped with an independent drive unit to receive power commands from the main control module and output the corresponding irradiance. The pose execution module is used to drive the segmented UVC lamp array module to adjust the irradiation pose and feed the real-time pose data back to the main control module. The light field calibration module is used to calibrate the light field contribution matrix of each lamp segment to each target surface unit under different illumination postures, and store the calibration data to the main control module. The constraint optimization solution module is used to receive the target surface unit parameters and the optical field contribution matrix, construct the constraint optimization model and solve the optimal power vector, and output it to the main control module. The pose planning module is used to identify undercovered targets based on the predicted dose under optimal power, and to plan a second compensation pose from the candidate pose set and output it to the main control module. The main control module is used to schedule each module to execute the shading compensation method for multi-angle segmented UVC irradiation as described in any one of claims 1-9.
