Device and method for self-adaptively adjusting power output of automatic driving tractor

Abstract

The invention discloses a device and a method for self-adaptively adjusting power output of an automatic driving tractor. Relates to the technical field of agricultural machinery automatic driving and power control. The rotary cultivator state sensing module is used for collecting real-time working state parameters of the rotary cultivator in real time; the central control module is used for generating a power adjustment instruction for adjusting the power output of the tractor and / or the working parameters of the rotary cultivator and an automatic driving correction instruction for correcting the driving state of the tractor based on land parameters and the real-time working state parameters of the rotary cultivator; the power adjusting module is used for adjusting power output of the tractor and / or working parameters of the rotary cultivator; and the automatic driving module is used for controlling the tractor to run according to the planned path and correcting the running state according to the automatic driving correction instruction so as to realize intelligent and self-adaptive regulation and control of power output of the automatic driving tractor in a complex and heterogeneous farmland operation environment.

Description

Technical Field

[0001] This invention relates to the field of automatic driving and power control technology for agricultural machinery, and in particular to a device and method for adaptively adjusting power output of an automatic driving tractor. Background Technology

[0002] Rotary tillage is a crucial step in agricultural production, and its quality directly affects the effectiveness of subsequent processes such as sowing and fertilization. With the development of agricultural mechanization and intelligence, self-driving tractors have been gradually applied to rotary tillage, significantly reducing the intensity of manual labor.

[0003] Currently, tractor systems employing autonomous driving technology typically consist of a path planning module, a power system, and implements (such as rotary tillers). During operation, the system primarily relies on preset programs and fixed power output modes, or can only be adjusted manually, lacking the ability to adaptively and finely adjust to real-time land conditions. However, farmland conditions are not uniform; soil hardness and moisture vary significantly across different locations in the field, and the distribution of residual mulch such as straw is uneven. These changing land conditions directly cause dynamic and drastic fluctuations in the workload of the rotary tiller.

[0004] When the soil in the work area is hard and the straw coverage is high, the fixed power output is difficult to meet the rotary tillage load requirements, which can easily lead to problems such as insufficient rotary tillage depth, incomplete straw crushing, and engine stalling or overloading due to insufficient tractor power, while also aggravating mechanical wear. When the soil is loose and the moisture is suitable, excessive power output will cause fuel / energy waste and may damage the soil structure due to excessive rotary tillage.

[0005] Furthermore, the lack of coordination between power regulation and automatic driving systems in existing technologies affects the consistency of rotary tillage operations. When an increased load is detected, a common strategy is to simply instruct the engine or drive motor to increase output power to overcome resistance. This piecemeal approach fails to consider the tractor's traveling power and the rotary tiller's working power as a coupled whole. The lack of coordinated optimization in the distribution of power between the vehicle's drive system and the implements can easily lead to low overall energy efficiency, or unreasonable power distribution under complex working conditions, resulting in decreased driving stability or uneven work quality.

[0006] To address these issues, there is an urgent need to develop a technical solution that can perceive land conditions in real time and adaptively match power output with the needs of rotary tillage operations. This solution should be combined with an autonomous driving system to form collaborative control, thereby improving the intelligence level, operational quality, and energy efficiency of rotary tillage operations. Summary of the Invention

[0007] Based on the above problems, the purpose of this invention is to provide a device and method for adaptively adjusting the power output of an autonomous driving tractor. By constructing a fully closed-loop control system of multi-source perception, intelligent decision-making and collaborative execution, intelligent and adaptive control of the power output of the autonomous driving tractor is realized in complex and heterogeneous farmland operation environments.

[0008] In a first aspect, the present invention provides a device for adaptively adjusting the power output of an autonomous driving tractor, comprising:

[0009] The land condition sensing module is used to collect at least one land parameter in the work area in real time.

[0010] The rotary tiller status sensing module is used to collect real-time working status parameters of the rotary tiller.

[0011] The central control module is communicatively connected to the land condition sensing module and the rotary tiller status sensing module. It is used to generate power adjustment commands for adjusting the tractor power output and / or the rotary tiller working parameters, and automatic driving correction commands for correcting the tractor driving status, based on the land parameters and the real-time working status parameters of the rotary tiller.

[0012] The power adjustment module is communicatively connected to the central control module and is used to adjust the power output of the tractor and / or the working parameters of the rotary tiller according to the power adjustment command.

[0013] The automatic driving module is communicatively connected to the central control module and is used to control the tractor to travel along the planned path, and can correct the driving status according to the automatic driving correction instructions.

[0014] Preferably, the land condition sensing module includes at least one of a first sensor for collecting soil hardness, a second sensor for collecting soil moisture, and a visual acquisition unit for identifying straw coverage; the rotary tiller status sensing module includes a torque sensor for collecting cutter shaft torque and / or a speed sensor for collecting cutter shaft rotation speed.

[0015] Preferably, the central control module processes the land parameters and real-time operating status parameters through an optimization model to generate the power adjustment command.

[0016] Furthermore, the optimization model is a multivariate coupled model, which is configured to: receive a comprehensive input vector containing land parameters and real-time working status parameters, and output a comprehensive output vector containing at least two kinds of dynamic parameters, wherein each dynamic parameter in the comprehensive output vector is determined collaboratively after coupled calculation based on the principle of minimizing the total energy consumption of the system.

[0017] Furthermore, the integrated input vector includes at least two of the following: soil hardness, soil moisture, straw coverage, real-time torque of the rotary tiller's cutter shaft, and real-time rotational speed; the integrated output vector includes the target output power of the tractor's range extender, the target power of the drive motor, and the target rotational speed of the rotary tiller's cutter shaft.

[0018] Preferably, the multivariate coupling model includes a load prediction sub-model for calculating the total demand load torque for rotary tillage operations; the load prediction sub-model collaboratively determines the total demand load torque by using the specific soil resistance coefficient based on land parameters and a dynamic compensation term based on real-time load feedback.

[0019] The soil impedance coefficient represents the load torque caused by the soil at a unit rotation speed. It is calculated by a nonlinear coupling model based on soil hardness, soil moisture, straw coverage rate and preset optimal cultivation moisture. The influence of soil moisture and straw coverage rate on load torque is modeled as a nonlinear function relationship.

[0020] The dynamic compensation term is calculated based on the real-time torque of the rotary tiller's cutter shaft, its filtered steady-state value, and the torque change rate.

[0021] Preferably, the multivariate coupling model further includes an optimized allocation sub-model;

[0022] The optimized allocation sub-model is configured to: with the goal of minimizing the total energy consumption of the system, based on the total demand load torque and the vehicle driving power demand determined by the autonomous driving module, perform collaborative optimization calculations on the range extender output power, drive motor power and rotary tiller target cutter shaft speed to generate the comprehensive output vector.

[0023] The principle for determining the target cutter shaft speed by the optimized allocation sub-model is: when the soil impedance coefficient increases, the target cutter shaft speed is automatically reduced.

[0024] Preferably, the optimized allocation sub-model is configured to: when the specific soil impedance coefficient is higher than a first preset threshold, output an instruction to increase the power of the range extender and / or drive motor while reducing the target cutter shaft speed; and when the specific soil impedance coefficient is lower than a second preset threshold, output an instruction to reduce the power of the range extender and / or drive motor while maintaining or increasing the target cutter shaft speed.

[0025] Preferably, the central control module is further configured to: compare the real-time torque of the rotary tiller's cutter shaft with a preset safe torque threshold in real time, and / or monitor whether the rate of change of the cutter shaft torque exceeds a preset abrupt change threshold; when the real-time torque of the cutter shaft exceeds the safe torque threshold, or the rate of change of torque exceeds the abrupt change threshold, generate an early warning command immediately.

[0026] Secondly, the present invention provides a method for adaptively adjusting the power output of an autonomous driving tractor, implemented by the apparatus described in the embodiments of the present invention, the method comprising:

[0027] Real-time collection of at least one land parameter in the work area, and simultaneous collection of real-time working status parameters of the rotary tiller;

[0028] Based on the land parameters and the real-time working status parameters of the rotary tiller, power adjustment commands are generated for adjusting the tractor's power output and / or the rotary tiller's working parameters, as well as automatic driving correction commands for correcting the tractor's driving status.

[0029] Adjust the tractor's power output and / or the rotary tiller's operating parameters according to the power adjustment command;

[0030] It controls the tractor to travel along the planned path and can correct the driving status according to the automatic driving correction instructions.

[0031] Compared with existing technologies, the beneficial effects of this invention include at least the following: By deeply integrating land condition sensing (soil hardness, moisture, straw coverage) with the real-time status sensing of the rotary tiller (cutter shaft torque, speed), this invention enables the system not only to sense the current load but also to understand the root cause of load changes (whether it is soil compaction, excessive moisture, or straw entanglement), and to make forward-looking predictions of the working resistance based on land parameters. Traditional methods often adjust engine power or working implement parameters in isolation. This invention, through a multivariate coupled optimization model of the central control module, takes minimizing the total system energy consumption as the global objective and performs integrated and coordinated optimization calculations on the range extender output power, drive motor power, and rotary tiller cutter shaft speed. Especially under high-resistance conditions, it adopts a coupled strategy of moderately reducing speed and coordinating power increase, rather than blindly increasing all power, scientifically balancing operational needs and energy consumption. This global coordinated optimization mechanism can tap into energy-saving potential at the system level, significantly reducing fuel or electricity consumption per unit area and directly improving operational economy. This device possesses dual adjustment capabilities: on the one hand, it precisely matches power and load through power adjustment commands to ensure uniform tillage depth and meet soil breaking quality standards, avoiding a decline in work quality due to insufficient or excessive power; on the other hand, it uses automatic driving correction commands to instantly compensate and correct vehicle driving status (such as speed) when load changes abruptly, effectively suppressing the interference of work load on the accuracy of automatic driving path tracking, ensuring smooth driving and trajectory accuracy. This power-driving dual closed-loop control enables the tractor to maintain high-quality and stable autonomous operation even in varied field conditions. It integrates safety monitoring and early warning functions based on real-time torque and torque change rate. When continuous mechanical overload or instantaneous rigid impact (such as hitting rocks) is detected, it can immediately trigger graded early warning commands (such as speed reduction, stopping, or vehicle avoidance), achieving proactive protection of core implements, effectively preventing mechanical damage, and extending equipment life. This safety fallback function, combined with the upper-level optimized control, forms a complete intelligent operating system that balances efficient operation and reliable performance. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of a device for adaptively adjusting power output of an autonomous driving tractor according to an embodiment of the present invention.

[0033] Figure 2 This is a schematic diagram of a method for adaptively adjusting the power output of an autonomous driving tractor according to an embodiment of the present invention. Detailed Implementation

[0034] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, they are provided to make the invention more comprehensive and complete, and to fully convey the concept of the exemplary embodiments to those skilled in the art. The same reference numerals in the drawings denote the same or similar structures, and therefore repeated descriptions of them will be omitted.

[0035] The terms used to express position and direction in this invention are illustrated with reference to the accompanying drawings, but changes can be made as needed, and all such changes are included within the scope of protection of this invention.

[0036] Example 1, refer to Appendix Figure 1 This embodiment provides a device for adaptively adjusting the power output of an autonomous driving tractor, comprising:

[0037] The land condition sensing module is used to collect at least one land parameter in the work area in real time.

[0038] The rotary tiller status sensing module is used to collect real-time working status parameters of the rotary tiller; these parameters include the cutter shaft torque and the operating speed.

[0039] Central control module: It is communicatively connected to the land condition perception module and the rotary tiller status perception module, and is used to generate power adjustment commands for adjusting tractor power output and / or rotary tiller working parameters, and automatic driving correction commands for correcting tractor driving status, based on the land parameters and the real-time working status parameters of the rotary tiller.

[0040] The power adjustment module is connected to the tractor's methanol range extender power system and includes a range extender power regulator, a motor controller, and a rotary tiller speed regulator. It is communicatively connected to the central control module and is used to adjust the tractor's power output and / or the rotary tiller's operating parameters according to the instructions of the central control module.

[0041] The automatic driving module is communicatively connected to the central control module and is used to control the tractor to travel along the planned path, and can correct the driving status according to the land parameters and / or the load data.

[0042] The land condition sensing module includes at least one of a first sensor for collecting soil hardness, a second sensor for collecting soil moisture, and a visual acquisition unit for identifying straw coverage; the rotary tiller status sensing module includes a torque sensor for collecting cutter shaft torque and / or a speed sensor for collecting cutter shaft rotation speed.

[0043] The first sensor is a penetrating pressure sensor with an adjustable measurement depth range of 5-20cm and a measurement accuracy error ≤ ±0.2MPa. The first sensor and the high-frequency capacitive soil moisture sensor are installed 50cm in front of the rotary tiller's blade shaft, with a measurement depth set to 15cm.

[0044] The visual acquisition unit is equipped with an AI image recognition algorithm to identify straw coverage rate; the straw coverage rate identification accuracy is ≥90%.

[0045] The visual acquisition unit includes a high-definition camera and an edge computing module. The high-definition camera is installed at the front or top of the tractor cab, facing forward of the work, and the shooting angle is adjustable. The edge computing module identifies the straw coverage rate through an image segmentation algorithm with an accuracy of ≥90%, and can eliminate the interference of soil shadows and weeds on the identification results. The data latency after processing is ≤100ms, ensuring the real-time performance and accuracy of the straw coverage rate parameter.

[0046] The torque sensor is a non-contact torque sensor installed at the input end of the rotary tiller's cutter shaft. It has a measurement range of 0-500 N·m and a measurement accuracy of ±1%FS. The collected torque data is transmitted synchronously with the speed data collected by the speed sensor, providing an accurate data source for load rate calculation.

[0047] In practical applications, during the preparation phase, operators set operating parameters such as rotary tillage depth and width through terminal equipment, and plan a complete operating path through the GPS positioning module. The central control module initializes the power output reference value according to the operating parameters, including the range extender reference power, the motor reference power, and the rotary tiller standard speed, to ensure that the power output is within a reasonable range when the operation starts.

[0048] As the tractor performs rotary tillage along the planned path, the land condition sensing module enters real-time data collection mode.

[0049] Its equipment includes a penetrating pressure sensor (first sensor) that measures soil hardness (in MPa) in real time at an adjustable depth (e.g., 15cm) in the area ahead or currently in operation; a humidity sensor (second sensor) simultaneously measures the moisture of the surface and topsoil; the sensors collect data every 0.5 seconds, and a vision acquisition unit, including a camera and equipped with an image recognition algorithm, captures and identifies images every 1 second to estimate the straw coverage rate of the surface area ahead. These land parameters together form the basis for predicting the resistance of the working environment. All data, along with GPS location information, is transmitted in real time to the central control module; simultaneously, the rotary tiller's status sensing module collects cutter shaft torque and speed data, providing feedback on the current operating load.

[0050] The central control module, as the core, simultaneously receives heterogeneous data streams from the two modules mentioned above. Instead of simply processing the two types of data independently, this module fuses and analyzes them to establish an intrinsic relationship between land conditions and the rotary tiller's operating status. Based on this comprehensive analysis, the central control module simultaneously generates two logically related commands:

[0051] a. Power Adjustment Command: This command aims to directly adjust the energy distribution and consumption of the operating system. Based on a comprehensive assessment of soil resistance and current load, it may instruct the tractor's power system (such as the engine or motor) to increase or decrease output power, and / or instruct the rotary tiller to adjust its operating parameters (such as cutter shaft speed) to proactively match or respond to changing working conditions.

[0052] b. Automated Driving Correction Instructions: These instructions aim to ensure the vehicle's driving stability and trajectory accuracy. Their generation logic considers the direct impact of land parameters on driving resistance (such as slipperiness) or changes in vehicle state that may be caused by power adjustments (such as brief lag due to sudden load increases), thereby making real-time fine adjustments to the basic path tracking control.

[0053] The power adjustment module receives and executes power adjustment commands, specifically adjusting the power output of the tractor and / or the operating parameters of the rotary tiller.

[0054] While controlling the vehicle to drive along a preset path, the autonomous driving module receives and executes autonomous driving correction commands to make real-time corrections to the driving status (such as speed and steering).

[0055] The two sensing modules continuously provide the latest land and state parameters, and the central control module updates its decisions accordingly, outputting new adjustment and correction commands, thereby enabling the tractor to continuously adaptively optimize power output and driving posture in heterogeneous fields.

[0056] The effects of the above technical solution are as follows:

[0057] By placing traditional vehicle driving control and the power control of the operating machinery within the same intelligent decision-making framework, and through the synchronous analysis and fusion of land parameters and real-time working status parameters of the rotary tiller by the central control module, the system can understand the root cause of changes in the operating load (whether it is soil hardening, changes in moisture, or an increase in straw), thereby making more reasonable global decisions. This breaks the previous situation of information isolation and fragmented control between the driving system and the operating system.

[0058] The system no longer passively responds to changes in the rotary tiller's load, but can also make a certain degree of feedforward predictions based on information about the land ahead. For example, when it senses an increase in straw coverage ahead, the central control module can generate an instruction to increase power in advance, making the power response more timely and effectively avoiding stalling or driving stagnation caused by sudden increases in resistance. This significantly improves the continuity and stability of operations in non-uniform fields.

[0059] By generating and executing automatic driving correction commands, this device incorporates the interference of workload on the vehicle's driving state into the scope of active control. When the rotary tiller's load increases, causing the vehicle to decelerate, the automatic driving module can promptly compensate and correct to maintain the set operating speed, thereby ensuring consistent tillage depth. Simultaneously, corrections for special terrain conditions such as slippery surfaces also ensure vehicle tracking accuracy and safety, enabling high-quality operation and precise automatic driving to be achieved simultaneously.

[0060] In one possible implementation, the central control module processes the land parameters and real-time operating status parameters through an optimization model to generate the power adjustment command.

[0061] The optimization model is a multivariate coupled model, which is configured to receive a comprehensive input vector containing land parameters and real-time working status parameters, and output a comprehensive output vector containing at least two kinds of dynamic parameters. The dynamic parameters in the comprehensive output vector are determined collaboratively after coupled calculation based on the principle of minimizing the total energy consumption of the system.

[0062] The comprehensive input vector includes: soil hardness H (unit: MPa), soil moisture M (percentage), straw coverage S (percentage), real-time torque T of the rotary tiller cutter shaft (unit: N·m) and real-time rotation speed. At least two of the following (units: rpm or rad / s); the integrated output vector includes: the target output power of the tractor range extender. (Unit: kW) Target power of the drive motor (Unit: kW) and the target cutter shaft speed of the rotary tiller (Unit: rpm or rad / s).

[0063] In one possible implementation, the device is further configured to preprocess land parameters and real-time operational status parameters, the preprocessing including at least one of the following steps:

[0064] The real-time torque of the rotary tiller's cutter shaft is filtered to extract its steady-state component;

[0065] Based on historical operation data or preset thresholds, abnormal input values ​​are eliminated or smoothed.

[0066] In one possible implementation, the multivariate coupled model includes a load prediction sub-model for calculating the total demand load torque for rotary tillage operations; the load prediction sub-model collaboratively determines the total demand load torque by using a specific soil resistivity coefficient based on land parameters and a dynamic compensation term based on real-time load feedback.

[0067] The soil resistance coefficient represents the load torque caused by the soil at a unit rotation speed. It is calculated by a nonlinear coupling model based on soil hardness, soil moisture, straw coverage, and a preset optimal cultivation moisture. The effects of soil moisture and straw coverage on load torque are both modeled as nonlinear functional relationships. For example, the effect of soil moisture on resistance is modeled as a Gaussian distribution centered on the optimal cultivation moisture, and the effect of straw coverage is modeled as a power function.

[0068] The dynamic compensation term is calculated based on the real-time torque of the rotary tiller's cutter shaft, its filtered steady-state value, and the torque change rate.

[0069] In one possible implementation, compared to the soil resistance coefficient Calculated using the following formula:

[0070]

[0071] Where H represents soil hardness and M represents soil moisture; The preset optimal soil moisture level; This indicates the nonlinear characteristic of resistance changing in a Gaussian distribution when humidity deviates from the optimal value; S is the straw coverage rate. The impact index of straw entanglement, and ; These are the model parameters calibrated using field data;

[0072] For example, preset optimal cultivation humidity The values ​​are 20%-30% (consistent with the unit of soil moisture); the straw entanglement influence index λ is 1.2-2.0 (calibrated through field tests based on straw toughness); α is 0.5-2.0 N·m / (rpm·MPa), β is 2.0-5.0, and γ is 0.3-1.0 N·m / rpm. The three values ​​are fitted and calibrated through field test data of at least 3 soil types (sandy soil, clay soil, loam) and 5 moisture gradients.

[0073] The load prediction sub-model is configured to calculate the total required load torque for rotary tillage operations using the following formula. :

[0074]

[0075] This represents the steady-state load torque caused by soil and crop residues; This represents the dynamic torque increment caused by real-time load fluctuations. Obtained using the following formula:

[0076]

[0077] T represents the real-time load torque of the rotary tiller shaft; This is the steady-state value of the real-time load torque of the rotary tiller shaft after filtering. The rate of change of the tool shaft torque; k, For dynamic compensation coefficients;

[0078] Wherein, the dynamic compensation coefficient k is a dimensionless coefficient, ranging from 0.1 to 0.3, and τ is in seconds, ranging from 0.1 to 0.5 seconds; the filtered steady-state torque The filter cutoff frequency is set to 5Hz using a first-order low-pass filter algorithm to filter sensor noise and instantaneous load disturbances.

[0079] The load prediction sub-model is configured to calculate the power required for rotary tillage (unit: kW) based on the total demand load torque and the target operating speed.

[0080] The process of generating the integrated output vector is configured to perform the following power balancing operation:

[0081]

[0082] The value for the rotary tiller's transmission system efficiency ranges from 0.85 to 0.95 (dynamically corrected based on the wear of the transmission mechanism).

[0083] 9550 is the conversion constant for speed-torque-power; (when Units are (Applicable when the unit is rpm).

[0084] The working principle of the above technical solution is as follows:

[0085] By establishing a load prediction sub-model that includes nonlinear coupling relationships, the system can quantitatively and accurately predict the upcoming operational load based on prior information such as soil hardness, moisture, and straw coverage, rather than simply performing lag compensation for the load that has already occurred. This achieves true feedforward control and significantly improves the system's response speed and smoothness.

[0086] The formula for calculating the soil resistivity coefficient K_soil is not a black box model; each term has a clear physical or agronomic meaning. For example, The term precisely describes the bell-shaped nonlinear relationship between soil moisture and resistance around the optimal tillage moisture; The term (λ>1) more accurately describes the phenomenon that increased straw coverage leads to a faster increase in resistance. This mechanism-based modeling enables the system to more intelligently understand the differentiated impact of different combinations of land conditions (such as hard and dry vs. soft but wet) on the load, thereby making more reasonable and adaptive control decisions.

[0087] By introducing a dynamic compensation term ΔT_dyn, the system can not only cope with the trend changes in land characteristics, but also quickly smooth out instantaneous impact loads caused by random factors (such as stones and tree roots). Through filtering and rate-of-change-based compensation, the interference of sudden load changes on the power system and driving stability is effectively suppressed, ensuring the continuity and high quality of operation, and reducing the impact on mechanical components.

[0088] Key parameters in the optimized model (such as α, β, γ, λ, k, and τ) were calibrated using field trial data, ensuring the model's applicability to different regions and crop residues. This data-driven parameterized model framework facilitates the subsequent integration of machine learning algorithms and the use of historical data for self-learning and optimization (such as updating). This (or by adjusting the λ value) laid a solid foundation, enabling the system to possess the intelligent potential for continuous evolution.

[0089] In one possible implementation, the multivariate coupling model further includes an optimization allocation sub-model;

[0090] The optimized allocation sub-model is configured to: with the goal of minimizing the total energy consumption of the system, based on the total demand load torque and the vehicle driving power demand determined by the autonomous driving module, perform collaborative optimization calculations on the range extender output power, drive motor power and rotary tiller target cutter shaft speed to generate the comprehensive output vector.

[0091] The principle for determining the target cutter shaft speed using the optimized allocation sub-model is: when the specific soil resistance coefficient increases, the target cutter shaft speed is automatically reduced.

[0092] The optimized allocation sub-model is configured to: when the specific soil impedance coefficient is higher than a first preset threshold, output an instruction to increase the power of the range extender and / or drive motor while reducing the target cutter shaft speed; when the specific soil impedance coefficient is lower than a second preset threshold, output an instruction to reduce the power of the range extender and / or drive motor while maintaining or increasing the target cutter shaft speed.

[0093] The optimized allocation sub-model receives two key inputs from the load forecasting sub-model: total demand load torque. and the power requirements for rotary tillage calculated therefrom And the vehicle's driving power requirements from the autonomous driving module. The model is solved with the goal of minimizing the total energy consumption of the system.

[0094] The model is based on total power demand. and consider Changes Feedback impact, dynamic collaborative allocation and ,make sure And all power components are operating within permissible limits.

[0095] To achieve this goal, its underlying decision-making logic is reflected in an intelligent trade-off strategy based on the specific soil resistance coefficient:

[0096] Strategies for reducing speed, maintaining torque, and increasing power in high-resistance scenarios: When When the speed exceeds a first preset threshold (e.g., conditions corresponding to hard soil, excessively wet / dry soil, or extremely high straw coverage), the model classifies it as a high-resistance scenario. The optimized comprehensive output vector will follow this principle: moderately reduce the target cutter shaft speed of the rotary tiller. This is because, under the same work quality requirements (e.g., tillage depth, soil breaking rate), appropriately reducing the speed in high-resistance soil can significantly reduce unnecessary power consumption required to overcome soil deformation and straw cutting (power is related to the cube or higher powers of the speed). Simultaneously, to compensate for insufficient traction or work intensity that may result from reduced speed, the model will collaboratively increase the range extender's output power and / or the drive motor's power to ensure that the total power is sufficient to maintain the set working speed and meet instantaneous torque requirements. This combination of reduction and increase represents a globally optimal solution based on the system's total energy consumption, rather than simply and crudely increasing power across the board.

[0097] Strategies for increasing engine speed, reducing power, and maintaining energy efficiency in low-resistance scenarios: When When the speed is below the second preset threshold (e.g., corresponding to loose soil, moderate moisture, and sparse straw), the model classifies it as a low-resistance scenario. In this case, maintaining or appropriately increasing the target rotor speed of the rotary tiller will not lead to a dramatic increase in power consumption; on the contrary, it may shorten the working time due to improved work efficiency, indirectly reducing energy consumption. Correspondingly, since the load itself is low, the model optimization calculation results in reducing the range extender output power and / or drive motor power, ensuring that the power output precisely matches the actual demand and avoiding energy waste.

[0098] By solving the above constrained optimization problem (which can usually be modeled as a nonlinear programming or model predictive control problem), the optimized allocation sub-model finally generates the comprehensive output vector: ( The three parameters in this vector are coupled, determined, and change collaboratively, together constituting the power distribution and operational parameter scheme that achieves maximum energy efficiency under specific land conditions and driving requirements. The central control module converts this vector into specific power adjustment commands and issues them for execution.

[0099] The effects of the above technical solution are as follows:

[0100] Unlike solutions that use a fixed rotation speed or simply adjust the rotation speed based on load feedback, this model actively adjusts the rotary tiller's rotation speed based on the specific soil resistance coefficient. It "actively reduces speed" under high resistance to avoid high-power consumption zones, and allows for speed increases under low resistance to improve efficiency. This adaptive speed management, aimed at optimizing energy efficiency, is key to reducing the core energy consumption of rotary tillage operations.

[0101] The model solves for the control variables of the range extender, drive motor, and rotary tiller speed within the same optimization framework. The output command is no longer the setpoint of three independent loops, but a collaborative optimal solution. For example, the commands to reduce speed and increase power under high resistance are calculated simultaneously and in the optimal ratio, ensuring that the system responds to changes in operating conditions in an overall optimal manner and avoiding overall energy efficiency losses caused by local optimization.

[0102] In areas with high straw coverage ( In large areas, a strategy of reducing rotational speed and increasing torque output provides more shear force, ensuring that the straw is thoroughly crushed rather than merely tangled, while maintaining a stable tillage depth. In soft soil, allowing for a slightly higher rotational speed improves soil pulverization and surface smoothness, enabling the device to not only save energy but also proactively optimize agronomic effects based on soil conditions.

[0103] The optimized allocation sub-model is calculated based on a continuous mathematical model and optimization algorithm. Its output is continuous and smoothly changing. Compared with action rules based on fixed thresholds and fixed proportions, it can avoid abrupt changes in instructions, making the power system and vehicle driving more stable, reducing mechanical shock and driver discomfort, and also better reflecting the actual situation of continuous parameter changes in complex farmland environments.

[0104] In one possible implementation, the optimized allocation sub-model has scenario classification priority logic, with the quantified load scenario judgment having a higher priority than the soil impedance coefficient threshold judgment. When the two scenario judgment results conflict, the adjustment logic corresponding to the quantified load scenario shall prevail.

[0105] The optimized allocation sub-model is set with emergency adjustment logic: when the rotary tiller cutter shaft torque exceeds the benchmark value by 30%, the energy consumption optimization target is paused, the range extender is prioritized to increase power by 1.5 times the load torque increment, the drive motor is supplemented at full power, and the rotary tiller cutter shaft speed is reduced to 60% of the rated speed.

[0106] The training data for the optimized allocation sub-model covers at least three soil types (sandy soil, clay soil, loam), five moisture gradients (10%-40%), three straw coverage gradients (20%, 40%, 70%), and two operating speeds (3km / h, 5km / h). At least 50 sets of valid data are collected for each scenario, and the optimal sample is labeled with the lowest energy consumption and the best operating quality (rotary tillage depth error ≤ ±5cm, straw crushing rate ≥ 85%).

[0107] The model parameter update mechanism of the central control module is as follows: after every 20 mu of work is completed, the update is triggered based on the deviation between the actual energy consumption and the model's predicted energy consumption. If the deviation is greater than 8%, the objective function weights are retrained, and the soil impedance coefficient threshold and load scenario judgment parameters are updated at the same time.

[0108] The optimized allocation sub-model supports transfer learning adaptation. For new operation areas, the input parameter weight coefficients can be fine-tuned by collecting 10 sets of pilot operation data without reconstructing the core operation logic, thus adapting to different regional soil characteristics and crop types.

[0109] The central control module is also configured to perform real-time safety monitoring and early warning based on the real-time working status parameters of the rotary tiller.

[0110] Specifically, the central control module compares the real-time torque of the rotary tiller's cutter shaft with a preset safe torque threshold in real time, and / or monitors whether the rate of change of the cutter shaft torque exceeds a preset sudden change threshold.

[0111] When the real-time torque of the cutter shaft exceeds the safe torque threshold, or the torque change rate exceeds the sudden change threshold, it is determined that the rotary tiller is facing an overload or an unpredictable rigid impact risk, and an early warning command is generated immediately.

[0112] The warning instruction includes at least one of the following:

[0113] Speed ​​reduction or shutdown command: sent to the power regulation module, requesting it to immediately reduce the rotary tiller's cutter shaft speed or perform an emergency shutdown;

[0114] Autonomous driving avoidance command: sent to the autonomous driving module, requesting it to control the tractor to slow down or stop moving.

[0115] The safe torque threshold and the mutation threshold can be determined based on the mechanical design limit of the rotary tiller and the theoretical load range under the current working conditions calculated by the load prediction sub-model, and can be dynamically adjusted according to the land parameters (such as soil hardness).

[0116] The working principle of the above technical solution is as follows:

[0117] By monitoring the real-time torque of the cutter shaft to ensure it does not exceed a safety threshold, the system can effectively identify the risk of continuous mechanical overload caused by extremely hard soil, the presence of hard objects underground, or improper operation. Simultaneously, by monitoring the rate of torque change to ensure it does not exceed a sudden change threshold, the system can sensitively detect instantaneous rigid impact events such as blade collisions with rocks or tree roots. This dual monitoring mechanism upgrades the traditional passive protection, which relies on operator experience, to proactive intelligent protection based on data feedback. This fundamentally avoids serious mechanical failures such as cutter shaft deformation and gearbox damage caused by overload or impact, significantly extending the service life of core operating equipment.

[0118] The warning commands generated by the system are not isolated actions, but rather form a coordinated control of the power and driving systems based on the risk level and type. For example, a sustained overload may trigger a command to reduce the rotary tiller's speed to alleviate the load; a severe impact may directly issue commands for emergency stop and tractor to halt movement. This tiered and coordinated response effectively mitigates risks while minimizing unnecessary interruptions to operational continuity, achieving an optimal balance between safety and operational efficiency.

[0119] Farmland environments are inherently unpredictable. This feature enables the system to handle unknown risks. When sensors detect abnormal loads (such as hidden large rocks) that cannot be predicted by soil parameters (such as conventional hardness and humidity), the safety monitoring mechanism, as a last line of defense, immediately activates to guide the system to avoid the danger. This significantly enhances the reliability and robustness of automated driving tractors in safe and autonomous operation in unstructured and heterogeneous field environments, reducing reliance on external monitoring or human intervention, and represents a crucial step towards fully autonomous operation.

[0120] By linking safety thresholds with the calculation results of the load prediction sub-model and real-time land parameters, the safety threshold is no longer a fixed value. For example, when the system predicts high-load conditions based on soil hardness, the safety torque threshold can be dynamically increased to avoid false alarms under normal high-load cultivation. In areas with loose soil, a more sensitive threshold for torque change rate can be set to prevent potential hidden obstacles. This dynamic adjustment capability ensures that safety monitoring is both sensitive and adaptable to different scenarios, reducing frequent false alarms or missed alarms caused by unreasonable threshold settings, and improving the overall intelligence level of the system and the user experience.

[0121] The central control module also has data storage and learning functions, which are used to update and optimize the parameters of the optimization model based on historical operation data.

[0122] Example 2, refer to Appendix Figure 2 This embodiment provides a method for adaptively adjusting the power output of an autonomous driving tractor, implemented using the device described in Embodiment 1. The method includes:

[0123] The land condition sensing module collects at least one land parameter in the work area in real time, and the rotary tiller status sensing module collects the real-time working status parameters of the rotary tiller in real time.

[0124] The central control module generates power adjustment commands for adjusting tractor power output and / or rotary tiller operating parameters, as well as automatic driving correction commands for correcting tractor driving status, based on the land parameters and the real-time working status parameters of the rotary tiller.

[0125] The power adjustment module adjusts the tractor's power output and / or the rotary tiller's operating parameters according to the power adjustment command.

[0126] The autonomous driving module controls the tractor to travel along the planned path and can correct the driving status according to the autonomous driving correction instructions.

[0127] During the preparation phase, the method further includes: planning the work path through the GPS positioning module, initializing the power output reference parameters through the central control module, and setting the standard operating speed of the rotary tiller and the reference values ​​for the tractor's travel speed.

[0128] The working principle and effect of the above technical solution are the same as those of Embodiment 1, and will not be repeated here.

[0129] This invention also provides a computer-readable storage medium for storing a computer program. When the computer program is executed, it implements the function of the device in Embodiment 1 of this invention. The specific implementation method and the technical effects achieved are the same as those described in the above method embodiments, and some details will not be repeated.

[0130] In this invention, a readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. The program product can take the form of any combination of one or more readable media. A readable medium can be a readable signal medium or a readable storage medium. A readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0131] Computer-readable storage media may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium capable of sending, propagating, or transmitting a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, or any suitable combination thereof. Program code for performing operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar programming languages. The program code may be executed entirely on a user computing device, partially on an associated device, as a standalone software package, partially on a user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing devices can be connected to user computing devices via any type of network, including local area networks (LANs) or wide area networks (WANs), or they can be connected to external computing devices (e.g., via the Internet using an Internet service provider).

[0132] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the invention without departing from the principles and spirit of the invention, and all such changes should fall within the protection scope of the claims of the present invention.

Claims

1. A device for adaptively adjusting power output of an automated driving tractor, characterized in that, include: The land condition sensing module is used to collect at least one land parameter in the work area in real time. The rotary tiller status sensing module is used to collect real-time working status parameters of the rotary tiller. The central control module is communicatively connected to the land condition sensing module and the rotary tiller status sensing module. It is used to generate power adjustment commands for adjusting the tractor power output and / or the rotary tiller working parameters, and automatic driving correction commands for correcting the tractor driving status, based on the land parameters and the real-time working status parameters of the rotary tiller. The power adjustment module is communicatively connected to the central control module and is used to adjust the power output of the tractor and / or the working parameters of the rotary tiller according to the power adjustment command. The automatic driving module is communicatively connected to the central control module and is used to control the tractor to travel along the planned path, and can correct the driving status according to the automatic driving correction instructions.

2. The apparatus according to claim 1, characterized in that, include: The land condition sensing module includes at least one of a first sensor for collecting soil hardness, a second sensor for collecting soil moisture, and a visual acquisition unit for identifying straw coverage; the rotary tiller status sensing module includes a torque sensor for collecting cutter shaft torque and / or a speed sensor for collecting cutter shaft rotation speed.

3. The apparatus according to claim 1, characterized in that, The central control module processes the land parameters and real-time operating status parameters through an optimization model to generate the power adjustment command.

4. The apparatus according to claim 1, characterized in that, The optimization model is a multivariate coupled model, which is configured to receive a comprehensive input vector containing land parameters and real-time working status parameters, and output a comprehensive output vector containing at least two kinds of dynamic parameters. The dynamic parameters in the comprehensive output vector are determined collaboratively after coupled calculation based on the principle of minimizing the total energy consumption of the system.

5. The apparatus according to claim 4, characterized in that, The comprehensive input vector includes at least two of the following: soil hardness, soil moisture, straw coverage, real-time torque of the rotary tiller's cutter shaft, and real-time rotational speed; the comprehensive output vector includes the target output power of the tractor's range extender, the target power of the drive motor, and the target rotational speed of the rotary tiller's cutter shaft.

6. The apparatus according to claim 5, characterized in that, The multivariate coupled model includes a load prediction sub-model for calculating the total demand load torque for rotary tillage operations. The load prediction sub-model determines the total demand load torque collaboratively by using the specific soil resistance coefficient based on land parameters and a dynamic compensation term based on real-time load feedback. The soil impedance coefficient represents the load torque caused by the soil at a unit rotation speed. It is calculated by a nonlinear coupling model based on soil hardness, soil moisture, straw coverage rate and preset optimal cultivation moisture. The influence of soil moisture and straw coverage rate on load torque is modeled as a nonlinear function relationship. The dynamic compensation term is calculated based on the real-time torque of the rotary tiller's cutter shaft, its filtered steady-state value, and the torque change rate.

7. The apparatus according to claim 6, characterized in that, The multivariate coupling model also includes an optimization allocation sub-model; The optimized allocation sub-model is configured to: with the goal of minimizing the total energy consumption of the system, based on the total demand load torque and the vehicle driving power demand determined by the autonomous driving module, perform collaborative optimization calculations on the range extender output power, drive motor power and rotary tiller target cutter shaft speed to generate the comprehensive output vector. The principle for determining the target cutter shaft speed by the optimized allocation sub-model is: when the soil impedance coefficient increases, the target cutter shaft speed is automatically reduced.

8. The apparatus according to claim 7, characterized in that, The optimized allocation sub-model is configured to: when the specific soil impedance coefficient is higher than a first preset threshold, output an instruction to increase the power of the range extender and / or drive motor while reducing the target cutter shaft speed; when the specific soil impedance coefficient is lower than a second preset threshold, output an instruction to reduce the power of the range extender and / or drive motor while maintaining or increasing the target cutter shaft speed.

9. The apparatus according to claim 3, characterized in that, The central control module is also configured to: compare the real-time torque of the rotary tiller's cutter shaft with a preset safe torque threshold in real time, and / or monitor whether the rate of change of the cutter shaft torque exceeds a preset abrupt change threshold; when the real-time torque of the cutter shaft exceeds the safe torque threshold, or the rate of change of the torque exceeds the abrupt change threshold, generate an early warning command immediately.

10. A method for adaptively adjusting power output of an automated driving tractor, characterized in that, Implemented by any one of the apparatuses described in claims 1-8, the method comprises: Real-time collection of at least one land parameter in the work area, and simultaneous collection of real-time working status parameters of the rotary tiller; Based on the land parameters and the real-time working status parameters of the rotary tiller, power adjustment commands are generated for adjusting the tractor's power output and / or the rotary tiller's working parameters, as well as automatic driving correction commands for correcting the tractor's driving status. Adjust the tractor's power output and / or the rotary tiller's operating parameters according to the power adjustment command; It controls the tractor to travel along the planned path and can correct the driving status according to the automatic driving correction instructions.

Images