Accurate deinsectization and weeding system and method based on unmanned aerial vehicle remote sensing and laser cooperation

Abstract

The invention discloses a precise deinsectization and weeding system and a precise deinsectization and weeding method based on unmanned aerial vehicle remote sensing and laser cooperation, and the key points of the technical scheme are that the precise deinsectization and weeding system based on unmanned aerial vehicle remote sensing and laser cooperation is provided; comprising an unmanned aerial vehicle platform, a remote sensing data acquisition module, a data processing and target identification module, a laser execution module, a positioning module and a central control module. The remote sensing data acquisition module, the data processing and target identification module, the laser execution module and the positioning module are fixed or integrated on the unmanned aerial vehicle platform and are electrically connected with the central control module; the remote sensing data acquisition module comprises a multispectral sensor and an infrared thermal imaging sensor; the laser execution module comprises a laser generator, a laser focal length adjusting assembly and a laser power adjusting assembly. According to the method, through an integrated framework of data integration, feature mining, model construction and quality monitoring, high efficiency of data integration is realized.

Description

Technical Field

[0001] This invention belongs to the field of remote sensing technology, specifically relating to a precision pest and weed control system and method based on UAV remote sensing and laser synergy. Background Technology

[0002] In agricultural production, weeds and pests are among the core issues affecting crop yield and quality. Traditional pest and weed control methods mainly rely on chemical spraying, which, while effective in the short term, has the following significant drawbacks: Severe environmental pollution: Chemicals can seep into the soil and groundwater, leading to soil compaction and water pollution. They also remain in crops, harming human health and failing to meet the needs of green agriculture development; Extremely low precision: Spraying is a "full coverage" method, unable to distinguish between crops, weeds, and pests, easily killing beneficial organisms (such as pollinating insects) and disrupting the ecological balance of farmland; Low operational efficiency: Manual operation of sprayers or large machinery is required, making operations difficult and costly in complex terrains (such as mountains and hills) or for tall crops (such as corn and sugarcane).

[0003] To address the aforementioned issues, existing technologies have developed drone-based pest and weed control solutions, which can be broadly categorized into two types: drone-based pesticide spraying, which optimizes operational efficiency but still relies on chemical pesticides, and the environmental pollution and residue problems remain unresolved.

[0004] Drone-based mechanical / laser weeding: Some solutions use drones equipped with mechanical cutting components or single laser devices, but they lack precise identification capabilities—they can only make a preliminary judgment of "foreign objects" through visual sensors, and cannot distinguish between weeds and crops, or pests and beneficial insects, which can easily cause accidental damage to crops. At the same time, these solutions do not combine the large-scale data collection capabilities of remote sensing technology, and can only achieve "point-to-point" local operations, which cannot meet the needs of efficient operations in large areas of farmland.

[0005] In addition, existing laser pest and weed control solutions have the problem of fixed laser parameters: different weeds (such as broadleaf weeds and grass weeds) have different leaf thicknesses and pests (such as aphids and cabbage caterpillars) have different body wall strengths. A laser with fixed power either cannot completely remove the target or damages the surrounding crops due to excessive power, resulting in extremely poor adaptability.

[0006] In summary, existing technologies lack a pest and weed control solution that can achieve coordinated "wide-range identification, precise positioning, and differentiated treatment," and cannot simultaneously meet the needs of green environmental protection, precision and efficiency, and wide adaptability. Summary of the Invention

[0007] In view of the problems mentioned in the background art, the purpose of this invention is to provide a precision pest and weed control system and method based on UAV remote sensing and laser collaboration, so as to solve the problems mentioned in the background art.

[0008] The above-mentioned technical objective of this invention is achieved through the following technical solution: a precision pest and weed control system based on UAV remote sensing and laser collaboration, comprising: a UAV platform, a remote sensing data acquisition module, a data processing and target recognition module, a laser execution module, a positioning module, and a central control module; the remote sensing data acquisition module, the data processing and target recognition module, the laser execution module, and the positioning module are all fixed or integrated into the UAV platform and are all electrically connected to the central control module; the remote sensing data acquisition module includes a multispectral sensor and an infrared thermal imaging sensor; the data processing and target recognition module has a built-in crop-weed-pest feature database and an AI recognition algorithm; the laser execution module includes a laser generator, a laser focal length adjustment component, and a laser power adjustment component; the positioning module adopts RTK positioning technology.

[0009] Preferably, the multispectral sensor is used to collect spectral data of farmland crops and weeds, and the spectral data covers the visible light band and the near-infrared band; the infrared thermal imaging sensor is used to collect thermal radiation data of pests on the crop surface.

[0010] Preferably, the AI ​​recognition algorithm is used to: distinguish between crops and weeds based on spectral data collected by a multispectral sensor by differences in spectral reflectance; identify pests based on thermal radiation data collected by an infrared thermal imaging sensor by differences in thermal radiation temperature; and output the location coordinates and type information of weeds and pests.

[0011] Preferably, the laser focal length adjustment component is used to adjust the position of the laser focus generated by the laser generator according to the flight altitude of the UAV platform, so that the laser focus falls on the surface of weeds or pests; the laser power adjustment component is used to adjust the output power of the laser generator according to the type of weeds or pests.

[0012] Preferably, the positioning module is used to obtain the real-time location coordinates of the UAV platform and convert the location coordinates of weeds or pests output by the data processing and target recognition module into absolute geographic coordinates.

[0013] This invention also discloses a precise pest and weed control method based on UAV remote sensing and laser collaboration, including the following steps: Step 1, Information collection: Input farmland boundary coordinates and crop type information to the central control module through the ground terminal, the central control module plans the flight path of the UAV, and the data processing and target recognition module loads the matching crop-weed-pest feature database.

[0014] Step 2, Data Transmission and Identification: The central control module controls the UAV platform to fly along the planned route, while the remote sensing data acquisition module simultaneously collects multispectral data and infrared thermal imaging data of farmland and transmits them to the data processing and target identification module.

[0015] Step 3, Algorithm Recognition: The data processing and target recognition module uses AI recognition algorithms to distinguish between crops and weeds, identify pests, and determine the type of weeds and pests; the positioning module converts the relative position of weeds or pests into absolute geographic coordinates and sends them to the central control module.

[0016] Step 4, Weeding Operation: The central control module controls the drone platform to fly to a preset height above the weeds or pests, sends parameter commands to the laser execution module, the laser focus adjustment component adjusts the laser focus to the target surface, the laser power adjustment component adjusts the laser power, and the laser generator emits laser to complete the pest and weed removal operation.

[0017] Step 5, Cyclic Weeding: The laser execution module sends a signal that the job is completed, the remote sensing data acquisition module collects data of the work area again, and the data processing and target recognition module verifies whether the target has been cleared; if not cleared, the laser power is adjusted and step 4 is repeated; if cleared, the UAV platform moves to the next area for operation.

[0018] Preferably, in step 1, the spacing of the UAV flight path is not greater than the effective detection width of the remote sensing data acquisition module.

[0019] Preferably, in step 4, the preset height range is 1.5m-3m, and the laser power adjustment range of the laser power adjustment component in step 4 is: 1W-3W for weeds and 5W-15W for pests.

[0020] In summary, the present invention has the following main advantages: The present invention achieves four core breakthroughs through an integrated architecture of "data integration - feature mining - model construction - quality monitoring": The present invention improves the efficiency of data integration: it constructs a unified paradigm and standard dataset for multi-heterogeneous power source data, is compatible with protocols such as DL / T645 and IEC61850, breaks through the "format barrier", and improves the data integration efficiency to more than 85%, supporting the global analysis of "source-grid-load".

[0021] The present invention improves the precision of feature extraction: it innovates a multi-dimensional time-series feature extraction algorithm to deeply mine "spatiotemporal coupling relationship" (such as regional power grid load transmission and dynamic correlation of multiple parameters of equipment) and "advanced semantic information" (equipment operating condition labels and operation mode classification), improving feature effectiveness by 40% and providing dual-dimensional support of "physical meaning + numerical law" for anomaly identification.

[0022] The present invention provides intelligent anomaly identification: it constructs an anomaly identification model based on robust extraction of spatiotemporal features and multi-scale temporal semantic modeling. Through robust extraction, harmonic noise and data missing interference are filtered out, reducing the false detection rate to below 15%. Through multi-scale semantic modeling, it adapts to data with different sampling frequencies, captures small sample / distribution drift anomalies, and reduces the missed detection rate to below 10%.

[0023] The quality monitoring closed-loop of this invention integrates a full-process power consumption data quality monitoring module to achieve a closed loop of "anomaly identification - root cause tracing - completion and repair". It has been verified in the production environments of ≥5 provincial companies, promoting the application of the technology to scenarios of "precise scheduling, equipment operation and maintenance, and demand response", and improving the efficiency of data quality governance by 60%. Attached Figure Description

[0024] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Example

[0026] refer to Figure 1 A precision pest and weed control system based on UAV remote sensing and laser collaboration includes: a UAV platform, a remote sensing data acquisition module, a data processing and target recognition module, a laser execution module, a positioning module, and a central control module; the remote sensing data acquisition module, the data processing and target recognition module, the laser execution module, and the positioning module are all fixed or integrated into the UAV platform and are all electrically connected to the central control module; the remote sensing data acquisition module includes a multispectral sensor and an infrared thermal imaging sensor; the data processing and target recognition module has a built-in crop-weed-pest feature database and an AI recognition algorithm; the laser execution module includes a laser generator, a laser focal length adjustment component, and a laser power adjustment component; the positioning module adopts RTK positioning technology.

[0027] The multispectral sensor is used to collect spectral data of crops and weeds in farmland, and the spectral data covers the visible light band and the near-infrared band; the infrared thermal imaging sensor is used to collect thermal radiation data of pests on the crop surface.

[0028] The AI ​​recognition algorithm is used to: distinguish between crops and weeds based on spectral data collected by multispectral sensors and differences in spectral reflectance; identify pests based on thermal radiation data collected by infrared thermal imaging sensors and differences in thermal radiation temperature; and output the location coordinates and type information of weeds and pests.

[0029] The laser focal length adjustment component is used to adjust the position of the laser focus generated by the laser generator according to the flight altitude of the UAV platform, so that the laser focus falls on the surface of weeds or pests; the laser power adjustment component is used to adjust the output power of the laser generator according to the type of weeds or pests.

[0030] The positioning module is used to obtain the real-time location coordinates of the UAV platform and convert the location coordinates of weeds or pests output by the data processing and target recognition module into absolute geographic coordinates.

[0031] This embodiment also discloses a precise pest and weed control method based on UAV remote sensing and laser collaboration, including the following steps: Step 1, Information collection: Input farmland boundary coordinates and crop type information to the central control module through the ground terminal, the central control module plans the UAV flight route, and the data processing and target recognition module loads the matching crop-weed-pest feature database.

[0032] Step 2, Data Transmission and Identification: The central control module controls the UAV platform to fly along the planned route, while the remote sensing data acquisition module simultaneously collects multispectral data and infrared thermal imaging data of farmland and transmits them to the data processing and target identification module.

[0033] Step 3, Algorithm Recognition: The data processing and target recognition module uses AI recognition algorithms to distinguish between crops and weeds, identify pests, and determine the type of weeds and pests; the positioning module converts the relative position of weeds or pests into absolute geographic coordinates and sends them to the central control module.

[0034] Step 4, Weeding Operation: The central control module controls the drone platform to fly to a preset height above the weeds or pests, sends parameter commands to the laser execution module, the laser focus adjustment component adjusts the laser focus to the target surface, the laser power adjustment component adjusts the laser power, and the laser generator emits laser to complete the pest and weed removal operation.

[0035] Step 5, Cyclic Weeding: The laser execution module sends a signal that the job is completed, the remote sensing data acquisition module collects data of the work area again, and the data processing and target recognition module verifies whether the target has been cleared; if not cleared, the laser power is adjusted and step 4 is repeated; if cleared, the UAV platform moves to the next area for operation.

[0036] In step 1, the spacing of the UAV flight path is not greater than the effective detection width of the remote sensing data acquisition module.

[0037] In step 4, the preset height ranges from 1.5m to 3m, and the laser power adjustment range of the laser power adjustment component in step 4 is 1W-3W for weeds and 5W-15W for pests.

[0038] This invention achieves four core breakthroughs through an integrated architecture of "data integration - feature mining - model construction - quality monitoring": First, it significantly improves data integration efficiency by constructing a unified paradigm and standard dataset for diverse and heterogeneous power source data, compatible with protocols such as DL / T645 and IEC61850, overcoming format barriers and increasing data integration efficiency to over 85%, supporting global analysis of "source-grid-load". Second, it enhances feature extraction accuracy by innovating a multi-dimensional time-series feature extraction algorithm, deeply mining "spatiotemporal coupling relationships" (such as regional power grid load transmission and dynamic correlation of multiple equipment parameters) and "advanced semantic information" (equipment operating condition labels and operating mode classification), improving feature effectiveness by 40% and providing dual-dimensional support of "physical meaning + numerical regularity" for anomaly identification. Third, it enables intelligent anomaly identification by constructing an anomaly identification model based on robust spatiotemporal feature extraction and multi-scale time-series semantic modeling: robust extraction filters harmonic noise and data missing interference, reducing the false detection rate to below 15%; multi-scale semantic modeling adapts to data with different sampling frequencies, capturing small sample / distribution drift anomalies, reducing the false detection rate to below 10%. The quality monitoring closed-loop of this invention integrates a full-process power consumption data quality monitoring module to achieve a closed loop of "anomaly identification - root cause tracing - completion and repair". It has been verified in the production environments of ≥5 provincial companies, promoting the application of the technology to scenarios of "precise scheduling, equipment operation and maintenance, and demand response", and improving the efficiency of data quality governance by 60%.

[0039] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A precision pest and weed control system based on UAV remote sensing and laser collaboration, characterized in that: include: The system includes an unmanned aerial vehicle (UAV) platform, a remote sensing data acquisition module, a data processing and target recognition module, a laser execution module, a positioning module, and a central control module. The remote sensing data acquisition module, data processing and target recognition module, laser execution module, and positioning module are all fixed to or integrated into the UAV platform and are all electrically connected to the central control module. The remote sensing data acquisition module includes a multispectral sensor and an infrared thermal imaging sensor. The data processing and target recognition module has a built-in crop-weed-pest feature database and AI recognition algorithm. The laser execution module includes a laser generator, a laser focus adjustment component, and a laser power adjustment component. The positioning module uses RTK positioning technology.

2. The precision pest and weed control system based on UAV remote sensing and laser synergy as described in claim 1, characterized in that, The multispectral sensor is used to collect spectral data of crops and weeds in farmland, and the spectral data covers the visible light band and the near-infrared band; the infrared thermal imaging sensor is used to collect thermal radiation data of pests on the crop surface.

3. The precision pest and weed control system based on UAV remote sensing and laser synergy as described in claim 1, characterized in that, The AI ​​recognition algorithm is used to: distinguish between crops and weeds based on spectral data collected by multispectral sensors and differences in spectral reflectance; identify pests based on thermal radiation data collected by infrared thermal imaging sensors and differences in thermal radiation temperature; and output the location coordinates and type information of weeds and pests.

4. The precision pest and weed control system based on UAV remote sensing and laser synergy as described in claim 1, characterized in that, The laser focal length adjustment component is used to adjust the position of the laser focus generated by the laser generator according to the flight altitude of the UAV platform, so that the laser focus falls on the surface of weeds or pests; the laser power adjustment component is used to adjust the output power of the laser generator according to the type of weeds or pests.

5. A precision pest and weed control system based on UAV remote sensing and laser synergy as described in claim 1, characterized in that, The positioning module is used to obtain the real-time location coordinates of the UAV platform and convert the location coordinates of weeds or pests output by the data processing and target recognition module into absolute geographic coordinates.

6. A precise pest and weed control method based on UAV remote sensing and laser synergy, characterized in that, Includes the following steps: Step 1: Information collection: Input farmland boundary coordinates and crop type information to the central control module through the ground terminal. The central control module plans the UAV flight path, and the data processing and target recognition module loads the matching crop-weed-pest feature database. Step 2, Data Transmission and Identification: The central control module controls the UAV platform to fly along the planned route, while the remote sensing data acquisition module simultaneously collects multispectral data and infrared thermal imaging data of farmland and transmits them to the data processing and target identification module. Step 3, Algorithm Identification: The data processing and target identification module uses AI identification algorithms to distinguish between crops and weeds, identify pests, and determine the type of weeds and pests; the positioning module converts the relative position of weeds or pests into absolute geographic coordinates and sends them to the central control module. Step 4, Weeding Operation: The central control module controls the drone platform to fly to a preset height above the weeds or pests, sends parameter commands to the laser execution module, the laser focus adjustment component adjusts the laser focus to the target surface, the laser power adjustment component adjusts the laser power, and the laser generator emits laser to complete the pest and weed removal operation. Step 5, Cyclic Weeding: The laser execution module sends a signal that the job is completed, the remote sensing data acquisition module collects data of the work area again, and the data processing and target recognition module verifies whether the target has been cleared; if not cleared, the laser power is adjusted and step 4 is repeated; if cleared, the UAV platform moves to the next area for operation.

7. A precise pest and weed control method based on UAV remote sensing and laser synergy as described in claim 6, characterized in that: In step 1, the spacing of the UAV flight path is not greater than the effective detection width of the remote sensing data acquisition module.

8. A precise pest and weed control method based on UAV remote sensing and laser synergy according to claim 6, characterized in that: In step 4, the preset height ranges from 1.5m to 3m, and the laser power adjustment range of the laser power adjustment component in step 4 is: 1W-3W for weeds and 5W-15W for pests.

Images