Agricultural equipment working condition self-adaptive control method based on edge calculation

Abstract

The invention provides an agricultural equipment working condition adaptive control method based on edge calculation. The method comprises the following steps: constructing a fragmented control strategy according to terrain type information, operation object information and agricultural equipment information in reality; constructing an agricultural geographic model according to the obtained terrain information and the operation object information; delimiting an operation range on the agricultural geographic model, and dividing a strategy configuration area according to terrain information and operation object information in the operation range; configuring a corresponding operation control strategy according to the terrain information, the operation object information and the agricultural equipment information corresponding to the strategy configuration area; and after an operation path is planned, verified, checked and adjusted, the agricultural machinery equipment is started to work, and real-time operation and working condition monitoring is carried out through the on-site man-machine interaction equipment. According to the method, the fragmented control strategy is constructed, the multiple strategy configuration areas are divided in advance, and the corresponding appropriate control strategies are configured for the different strategy configuration areas, so that the actual working condition of the agricultural equipment is more adaptive to the real environment.

Description

Technical Field

[0001] The present invention relates to the field of agricultural machinery, and in particular to an adaptive control method for operating conditions of agricultural machinery based on edge computing. Background Art

[0002] In recent years, with the advancement of modern technology, the level of intelligence and automation in my country's agricultural machinery has continued to improve, and unmanned agricultural machinery has been widely adopted in actual agricultural operations. Automatic field navigation and automatic control of operating mechanisms are two fundamental functions of intelligent agricultural machinery. Automatic navigation primarily utilizes the machinery's machine vision and positioning systems to achieve autonomous sensing, positioning, and motion control. Automatic control of operating mechanisms relies on monitoring data from various sensors and real-time adjustments to the operating mechanism based on specific operational requirements and agronomic standards. The combination of these two functions ultimately enables high-quality and efficient automated navigation and operation. Unmanned agricultural machinery enables unmanned, automated operation of agricultural machinery, reducing manpower input and labor costs, and further improving agricultural production efficiency.

[0003] However, existing unmanned agricultural machinery generally uses general control strategies and cloud computing. However, in reality, field terrain is complex, and cloud computing has long latency, leading to frequent errors in unmanned agricultural machinery. General control strategies typically adjust associated parameters based on changes in certain operating parameters. This process is gradual or linear, making it difficult to adapt to sudden changes. For example, if crops suddenly fall, such as wheat, it can be difficult to adjust in a timely manner, making the fallen crops difficult to harvest or even crushed into the soil. Some control strategies even lack response strategies for such emergencies, resulting in poor adaptive operation results for agricultural machinery. Summary of the Invention

[0004] In order to solve the problems existing in the background technology, the present invention proposes an adaptive control method for working conditions of agricultural machinery based on edge computing.

[0005] The method for adaptive control of agricultural machinery working conditions based on edge computing includes the following steps:

[0006] S100, constructing a control strategy library, wherein the control strategy library includes multiple groups of fragmented control strategies constructed based on real terrain type information, operation object information, and agricultural machinery equipment information;

[0007] S200, obtaining terrain information and operation object information within a certain range;

[0008] S300, constructing an agricultural geographical model based on the acquired terrain information and operation object information;

[0009] S400, defining an operation range on an agricultural geographic model, and dividing a strategy configuration area within the operation range according to terrain information and operation object information within the operation range;

[0010] S500: Based on the terrain information, operation object information, and agricultural machinery information corresponding to the strategy configuration area, a corresponding operation control strategy is called from the control strategy library and configured;

[0011] S600: After planning the operation path and performing verification, checking and adjustment, the agricultural machinery equipment is started and the operation and working condition are monitored in real time through the on-site human-machine interaction equipment.

[0012] Based on the above, in step S600, after the operation path is planned, a pre-adjustment area is set in the current-level policy configuration area corresponding to the next-level policy configuration area in the direction of the operation path, and a transition control strategy is configured in the pre-adjustment area.

[0013] Based on the above, the terrain information at least includes flat land terrain information, slope and slope information, slope length information and terrain anomaly information.

[0014] Based on the above, the operation object information at least includes geological information, soil moisture information, operation requirement information, crop information and crop abnormality information.

[0015] Based on the above, the agricultural machinery information at least includes agricultural machinery type information, agricultural machinery model information and the operating condition information of each model of agricultural machinery under each type.

[0016] Based on the above, in step S200, aerial survey is performed using a drone and a total station to obtain terrain information, a terrain model is constructed based on the terrain information, and an agricultural geographical model is constructed based on the terrain model and the operation object information.

[0017] Based on the above, in step S600, by constructing constraint conditions, constraint verification is performed on the changes in the control strategy in the planned path direction and the associated corresponding terrain information, work object information and agricultural machinery equipment information.

[0018] Based on the above, in step S600, when planning the operation path, avoidance planning is performed for abnormal terrain; after avoidance planning is performed for abnormal crop areas, manual planning processing is performed on the abnormal crop areas.

[0019] Compared with the existing technology, the present invention has outstanding substantial features and significant progress. Specifically, the present invention constructs a fragmented control strategy, and after building an agricultural geographical model according to the terrain, work objects, and agricultural machinery and equipment, it pre-divides multiple strategy configuration areas, configures corresponding appropriate control strategies for different strategy configuration areas, and finally plans the work path. On the one hand, the work area is closer to the real environment. On the other hand, the area and control strategy are subdivided in advance, so that the agricultural machinery and equipment can fully adapt to the real environment when working on the planned path, so that the actual working conditions of the agricultural machinery and equipment are more compatible with the real environment, and it has the advantages of strong adaptability and easy use. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic block diagram of the process of the present invention. DETAILED DESCRIPTION

[0021] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without creative work are within the scope of protection of the present invention.

[0022] like Figure 1 As shown, the method for adaptive control of agricultural machinery working conditions based on edge computing includes the following steps: S100, constructing a control strategy library, which includes multiple groups of fragmented control strategies constructed according to actual terrain type information, work object information and agricultural machinery information; S200, obtaining terrain information and work object information within a certain range; S300, constructing an agricultural geographic model based on the acquired terrain information and work object information; S400, defining the working range on the agricultural geographic model, and dividing the strategy configuration area within the working range according to the terrain information and work object information within the working range; S500, calling and configuring the corresponding operation control strategy from the control strategy library according to the terrain information, work object information and agricultural machinery information corresponding to the strategy configuration area; S600, planning the working path and performing verification, checking and adjustment, starting the agricultural machinery and performing real-time operation and working condition monitoring through on-site human-computer interaction equipment.

[0023] Specifically, edge computing is to provide edge intelligent services nearby at the edge of the network close to the source of objects or data through a distributed open platform that integrates network, computing, storage, and application core capabilities. Simply put, edge computing is to analyze the data collected from the terminal directly on the local device or network close to the data generation, without the need to transmit the data to the cloud data processing center. Therefore, the data processing in this embodiment adopts the edge computing method, and is processed on the agricultural machinery itself, such as a tractor, or on a handheld interactive device such as a tablet computer, so as to improve the timeliness of data processing and transmission and avoid delays.

[0024] The terrain information at least includes flat terrain information, slope and slope information, slope length information and terrain anomaly information. The operation object information at least includes geological information, soil moisture information, operation requirement information, crop information and crop anomaly information. In reality, with the development of science and technology and policies, smart agriculture is now relatively common, so various monitoring data of farmland such as soil quality, soil moisture, crop rotation, etc. are recorded and can be directly obtained. The visual system, positioning system and various sensor devices for monitoring their own working conditions of agricultural machinery equipment are all possessed by unmanned agricultural machinery equipment itself. Therefore, the agricultural machinery equipment information at least includes agricultural machinery equipment type information, agricultural machinery equipment model information and the own working condition information of each model of agricultural machinery equipment under each type, etc., which are also convenient and direct to obtain. In this embodiment, terrain information is obtained through aerial survey using equipment such as drones and total stations, and a terrain model is constructed based on the terrain information (aerial survey modeling using drones and total stations is a common existing method, such as for map drawing and updating). Based on the terrain model, the terrain model is modeled and attributes are set based on the geology, soil moisture, crop type, crop height, and other information contained in the work object information, thereby constructing an agricultural geography model. It should be noted that the actual terrain is relatively complex, and some terrain anomalies exist, such as deep pits, rocks, and even gullies in some plots, as well as crop anomalies such as crop lodging in some areas. Therefore, the terrain model and agricultural geography model need to be manually checked and adjusted after construction.

[0025] In this embodiment, the fragmented control strategy refers to configuring a set of control strategies based on each type of terrain, each operating object, and each type of agricultural machinery information in reality. In other words, a fragmented subdivision is performed. For example, for slope terrain, the slope is divided into five types according to technical regulations, but the slope length is also taken into account. For example, when the slope is 2°, the drop length of a 50-meter slope reaches nearly 2 meters. Therefore, for different terrains and terrain conditions such as length, each operating object (such as harvesting, rotary tillage, etc.) and the operating object (such as wheat, corn, or rotary tillage soil hardness) are mapped one by one to each operating object under different terrains and terrain conditions such as length. Then, a one-to-one mapping is performed based on the model of agricultural machinery and the corresponding working conditions, thereby forming a subdivided fragmented control strategy. This is different from the existing method of setting a universal control strategy based on a single terrain information and operating object. For example, based on slope and rotary tillage alone, a universal control strategy corresponding to slope and rotary tillage is set, and then real-time calculation and adjustment are performed according to the real-time changes in working conditions. This method has a large amount of calculation and poor adaptability to the real environment.

[0026] Because aerial terrain surveys cover a large area, the constructed terrain model can be stored and reused later, requiring only re-surveys after a certain period of time to update any changes. However, the operation scope is typically limited to a specific area. For example, some fields may be planted with corn, while others may be planted with grain, requiring separate harvesting procedures. Therefore, the operation scope is manually defined within the constructed agricultural geographic model. Within the operation scope, regions are divided based on terrain information, such as flat areas and sloping areas, into different strategy configuration areas. Based on information such as the terrain, the operation target, and the agricultural machinery in question, matching fragmented control strategies are deployed from the control strategy library. Once strategies are configured for different areas within the operation scope, path planning for the agricultural machinery is performed based on the terrain and the machinery's own parameters. This includes planning for corners, unusual terrain, and unusual crop conditions. For example, areas with unusual crop conditions, such as fallen crops, are avoided, and targeted planning and harvesting operations are performed on fallen areas after all harvesting is complete. Thanks to the detailed configuration of control strategies, real-time adjustments to agricultural machinery in response to changes in operating conditions generally require only minor adjustments, requiring minimal computational effort. This significantly improves the efficiency of operating condition adjustments. Once path planning is complete, manually verified, and adjustments are made, the machinery can be activated and put into operation. Simultaneously, on-site human-computer interaction terminals, such as tablet computers, provide real-time monitoring of operating conditions, maximizing the adaptability of the machinery while ensuring safety.

[0027] In reality, before manual verification after path planning, constraints are constructed, such as constraining the calculation of each matching control strategy through the existing minimum constraint algorithm, so as to perform constraint verification on the changes in the control strategy in the planned path direction and its associated corresponding terrain information, work object information and agricultural machinery equipment information, so that the control strategy and its associated corresponding terrain information, work object information and agricultural machinery equipment information as well as the planned path, etc., can reach the optimal configuration as much as possible, so that the real-time working conditions of the agricultural machinery equipment are adjusted as little as possible relative to the control strategy, which is convenient for improving the adaptive efficiency of the agricultural machinery equipment.

[0028] Preferably, after the work path is planned, since the agricultural machinery may pass through different strategy configuration areas while moving along the planned path, the control strategies in different strategy configuration areas may vary to a certain extent. Therefore, along the work path direction, within the current level strategy configuration area, close to the position of the next configuration strategy area, a pre-adjustment area is set corresponding to the next level strategy configuration area, and a transition control strategy is configured in the pre-adjustment area. That is, the transition control strategy is a control strategy used to transition from the current level control strategy to the next area control strategy. For example, when transitioning from a flat area to a slope with a higher slope, a certain degree of throttle or even downshifting is performed in advance to avoid sudden and large-scale adjustments to the working condition parameters after entering the next area, thereby further improving the working condition adaptability of the agricultural machinery.

[0029] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. The method of adaptive control of agricultural machinery working conditions based on edge computing is characterized by: Including steps: S100, constructing a control strategy library, wherein the control strategy library includes multiple groups of fragmented control strategies constructed based on real terrain type information, operation object information, and agricultural machinery equipment information; S200, obtaining terrain information and operation object information within a certain range; S300, constructing an agricultural geographical model based on the acquired terrain information and operation object information; S400, defining an operation range on an agricultural geographic model, and dividing a strategy configuration area within the operation range according to terrain information and operation object information within the operation range; S500: Based on the terrain information, operation object information, and agricultural machinery information corresponding to the strategy configuration area, a corresponding operation control strategy is called from the control strategy library and configured; S600: After planning the operation path and performing verification, checking and adjustment, the agricultural machinery equipment is started and the operation and working condition are monitored in real time through the on-site human-machine interaction equipment.

2. The method for adaptive control of agricultural machinery operating conditions based on edge computing according to claim 1 is characterized in that: In step S600, after the operation path is planned, a pre-adjustment area is set in the direction of the operation path within the current-level policy configuration area corresponding to the next-level policy configuration area, and a transition control strategy is configured in the pre-adjustment area.

3. The method for adaptive control of agricultural machinery operating conditions based on edge computing according to claim 1 is characterized in that: The terrain information at least includes flat land terrain information, slope and slope information, slope length information and terrain anomaly information.

4. The method for adaptive control of agricultural machinery operating conditions based on edge computing according to claim 1 is characterized in that: The operation object information includes at least geological information, soil moisture information, operation requirement information, crop information and crop abnormality information.

5. The method for adaptive control of agricultural machinery operating conditions based on edge computing according to claim 1 is characterized in that: The agricultural machinery information includes at least agricultural machinery type information, agricultural machinery model information and the working condition information of each model of agricultural machinery under each type.

6. The method for adaptive control of agricultural machinery operating conditions based on edge computing according to claim 1, characterized in that: In step S200, aerial survey is performed using a drone and a total station to obtain terrain information, a terrain model is constructed based on the terrain information, and an agricultural geographical model is constructed based on the terrain model and the operation object information.

7. The method for adaptive control of agricultural machinery operating conditions based on edge computing according to claim 1, characterized in that: In step S600, by constructing constraint conditions, constraint verification is performed on the changes in the control strategy in the planned path direction and the corresponding terrain information, work object information and agricultural machinery equipment information.

8. The method for adaptive control of agricultural machinery operating conditions based on edge computing according to claim 1, characterized in that: In step S600, when planning the operation path, avoidance planning is performed for abnormal terrain; after avoidance planning is performed for abnormal crop areas, manual planning processing is performed on the abnormal crop areas.

Images