Transmission smooth transition control method and system for obstacle avoidance of unmanned agricultural machine

Abstract

The invention discloses a transmission smooth transition control method and system for obstacle avoidance of an unmanned agricultural machine, and relates to the technical field of obstacle avoidance control. The method comprises the steps that the current slip rate of each driving wheel is calculated based on vehicle state data collected in real time, and the current ground attachment coefficient is updated in real time; calculating a transverse deviation and a course deviation based on the current vehicle position and the course angle and a pre-loaded target path; based on the lateral deviation, the course deviation, the vehicle state data and the environment information, a preliminary control instruction is obtained through calculation; based on the current slip rate and the current ground attachment coefficient of each driving wheel, correcting a longitudinal control instruction in the preliminary control instruction to obtain a corrected longitudinal control instruction; based on the vehicle state data, correcting a transverse control instruction in the preliminary control instruction to obtain a corrected transverse control instruction; according to the invention, the agricultural machine can realize stable obstacle avoidance in a complex farmland environment, and the safety and reliability of the obstacle avoidance operation of the agricultural machine are improved.

Description

Technical Field

[0001] This invention relates to the field of obstacle avoidance control technology, and more specifically, to a transmission smooth transition control method and system for obstacle avoidance in unmanned agricultural machinery. Background Technology

[0002] In the actual operation of unmanned agricultural machinery, farming activities such as tilling, sowing, and spraying often need to be carried out in complex terrain environments. In such environments, the soil moisture and firmness are unevenly distributed, and terrain features such as furrows and field ridges are widespread, causing the adhesion coefficient between the ground and the driving wheels of the agricultural machinery to be constantly in a state of dynamic fluctuation. Current obstacle avoidance transmission control technology for unmanned agricultural machinery mostly simplifies the obstacle avoidance braking process into a one-dimensional longitudinal dynamics problem. Its core technical principle revolves around the smooth control of longitudinal vehicle speed and acceleration. Impact suppression during longitudinal deceleration is achieved by adjusting the engine output torque, gearbox transmission ratio, or braking force of the braking system. The control logic mainly relies on the feedback adjustment of longitudinal motion parameters such as vehicle speed and engine speed to achieve the goal of smooth transmission transition.

[0003] However, due to the dynamic changes in the coefficient of adhesion of farmland, the drive wheels of agricultural machinery are prone to sudden slippage during obstacle avoidance braking or torque adjustment. Changes in slippage state directly affect the force transmission characteristics of the steering system, which in turn causes changes in the yaw motion of the agricultural machinery. This correlation between slippage and attitude dynamics will further cause deviations between the actual trajectory of the agricultural machinery and the preset obstacle avoidance path. Existing technologies do not incorporate dynamic parameters such as slippage state and yaw motion into the scheduling logic of transmission control, and longitudinal torque smoothing alone cannot offset the interference caused by the coupling of the above-mentioned multiple factors.

[0004] This technical defect directly causes agricultural machinery to deviate from the preset path during obstacle avoidance, even if the longitudinal deceleration process remains stable. This makes it impossible for the machinery to accurately avoid obstacles, leading to scraping or even collisions. Furthermore, the deviation from the trajectory disrupts the continuity of the work path, causing problems such as uneven row spacing and spraying range deviation, resulting in a decline in work quality. In work scenarios where the ground adhesion coefficient fluctuates more significantly, the probability and severity of the above problems will be further amplified, affecting the operational safety and reliability of unmanned agricultural machinery.

[0005] In view of this, the present invention proposes a transmission smooth transition control method and system for obstacle avoidance in unmanned agricultural machinery to solve the above problems. Summary of the Invention

[0006] To overcome the aforementioned deficiencies of the prior art and achieve the above objectives, the present invention provides the following technical solution: a transmission smooth transition control method for obstacle avoidance in unmanned agricultural machinery, comprising:

[0007] Based on real-time collected vehicle status data, the current slip ratio of each drive wheel is calculated, and the current ground adhesion coefficient is updated in real time.

[0008] Based on the current vehicle position and heading angle, as well as the pre-loaded target path, calculate the lateral deviation and heading deviation;

[0009] Based on lateral deviation, heading deviation, vehicle status data, and environmental information, preliminary control commands are calculated.

[0010] Based on the current slip ratio and current ground adhesion coefficient of each drive wheel, the longitudinal control command in the initial control command is corrected to obtain the corrected longitudinal control command.

[0011] Based on vehicle status data, the lateral control command in the initial control command is corrected to obtain the corrected lateral control command.

[0012] The modified longitudinal control command and the modified lateral control command are used as the final control commands to drive the vehicle.

[0013] Furthermore, by using the linear velocity corresponding to the wheel speed of the drive wheel and the actual speed of the vehicle, the current slip ratio of each drive wheel is calculated; based on the current slip ratio of the drive wheels on the left and right sides of the vehicle and the preset difference threshold, it is determined whether there is a unilateral slip difference between the drive wheels on the left and right sides.

[0014] Furthermore, the engine output torque is converted into theoretical acceleration and compared with the actual longitudinal acceleration measured by the IMU to obtain the current ground adhesion coefficient.

[0015] Furthermore, if there is a difference in slippage on one side of the drive wheels on the left and right sides, it is determined whether the slippage on one side is a local working condition.

[0016] If it is determined to be a local working condition, the current overall ground adhesion coefficient is used as the benchmark, and the adhesion coefficient corresponding to the side with the higher slip ratio value of the left and right drive wheels is weighted and corrected. The weighting coefficient is determined according to the duration of slippage.

[0017] If the condition is determined to be non-local, the adhesion coefficient corresponding to the slip ratio of the drive wheel with the higher slip ratio value on the left and right sides shall be used as the current ground adhesion coefficient.

[0018] Furthermore, the method for modifying the longitudinal control command in the initial control command is as follows: determine the maximum allowable acceleration based on the current ground adhesion coefficient.

[0019] Furthermore, methods for determining the maximum permissible acceleration based on the current ground adhesion coefficient include:

[0020] The maximum longitudinal acceleration is calculated based on the current ground adhesion coefficient and gravitational acceleration.

[0021] If the target acceleration exceeds the maximum longitudinal acceleration, the target acceleration will be reduced to the maximum longitudinal acceleration.

[0022] Furthermore, during the braking process, if a drive wheel lock-up is detected, the pressure supply from the master cylinder to the corresponding drive wheel brake cylinder is reduced, and the opening of the brake pressure regulating valve is adjusted according to the drive wheel speed to reduce the hydraulic pressure value in the brake brake cylinder until the drive wheel speed matches the actual vehicle speed, thus releasing the drive wheel lock-up.

[0023] Furthermore, when the yaw rate change exceeds the preset yaw rate change threshold, or the absolute difference between the slip rates of the left and right drive wheels exceeds the preset slip rate difference threshold, the lateral control command in the initial control command is corrected.

[0024] Furthermore, if the actual yaw rate exceeds the expected yaw rate threshold calculated based on the current steering command, the engine torque output is reduced to a preset safe torque range.

[0025] Calculate the absolute difference between the slip ratios of the left and right drive wheels and compare it with a preset slip ratio difference threshold. If the difference exceeds the preset slip ratio difference threshold, the side with the higher slip ratio value of the left and right drive wheels is determined to be the side with severe slippage.

[0026] Apply single-sided braking to the drive wheel on the side with the most severe slippage;

[0027] When one drive wheel spins freely and causes the vehicle to yaw in the opposite direction, brakes are applied to that drive wheel to slow it down and correct the vehicle's yaw direction toward the target path.

[0028] When a vehicle is detected to have deviated from its path due to slippage, the steering angle command is increased to compensate for the path deviation, and the increase in the steering angle command does not exceed a preset compensation threshold.

[0029] The smooth transmission transition control system for obstacle avoidance in unmanned agricultural machinery includes:

[0030] The data calculation module calculates the current slip ratio of each drive wheel based on real-time collected vehicle status data and updates the current ground adhesion coefficient in real time.

[0031] The deviation calculation module calculates the lateral deviation and heading deviation based on the current vehicle position and heading angle, as well as the pre-loaded target path.

[0032] The command calculation module calculates preliminary control commands based on lateral deviation, heading deviation, vehicle status data, and environmental information.

[0033] The first command correction module corrects the longitudinal control command in the initial control command based on the current slip ratio and the current ground adhesion coefficient of each drive wheel, and obtains the corrected longitudinal control command.

[0034] The second instruction correction module corrects the lateral control instruction in the initial control instruction based on vehicle status data, and obtains the corrected lateral control instruction.

[0035] The command-driven module is used to drive the vehicle by taking the modified longitudinal control command and the modified lateral control command as the final control command.

[0036] Compared with the prior art, the technical effects and advantages of the transmission smooth transition control method and system for obstacle avoidance of unmanned agricultural machinery of the present invention are as follows:

[0037] The unmanned agricultural machinery obstacle avoidance transmission smooth transition control method and system of the present invention first collects vehicle status data in real time, including vehicle position, heading angle, yaw rate, and drive wheel speed, calculates the current slip ratio of each drive wheel and updates the ground adhesion coefficient in real time. The slip ratio calculation near zero speed is adjusted by setting a small threshold based on the agricultural machinery model and operating scenario and performing adaptive correction. In the event of unilateral slippage, terrain perception information, slippage duration, and changes in drive torque are used to distinguish between local and non-local operating conditions, and the ground adhesion coefficient is adjusted accordingly. Then, based on the current vehicle position, heading angle, and pre-loaded target path, lateral deviation and heading deviation are calculated. Finally, combining the lateral deviation, heading deviation, vehicle status data, and environmental information, preliminary control commands are generated to initiate steering. The control system combines vehicle speed, target path curvature, and real-time ground adhesion coefficient to adapt proportional control gain, while incorporating look-ahead distance to compensate for steering actuator response delay. Longitudinal control adjusts the target speed based on obstacle distance. Then, based on drive wheel slip ratio and ground adhesion coefficient, the longitudinal control command is corrected, limiting the maximum acceleration to match ground adhesion capability. During braking, the drive wheels are unlocked by adjusting the hydraulic pressure of the brake calipers. Simultaneously, lateral control commands are corrected based on vehicle yaw and unilateral slippage, correcting vehicle yaw through unilateral braking and traction distribution adjustment. When deviating from the path, the steering angle can be appropriately increased with a limited correction range. Finally, the corrected control commands are sent to the actuators, and closed-loop feedback continuously adjusts the control inputs to ensure precise command execution.

[0038] This invention changes the traditional obstacle avoidance method, which focuses solely on longitudinal dynamics control. It effectively solves the problems of existing technologies where, due to the lack of consideration of slippage and yaw motion parameters, the longitudinal deceleration of agricultural machinery is smooth, but the front of the machine easily deviates from the preset path, easily scrapes or even collides with obstacles. It also solves the drawbacks of trajectory deviation disrupting the continuity of the work path, causing uneven row spacing and spray range deviation, leading to a decline in work quality. Furthermore, it improves the situation of power waste and reduced work efficiency caused by the failure to distinguish between unilateral slippage conditions. Its advantages include enabling agricultural machinery to achieve stable obstacle avoidance in complex farmland environments with fluctuating soil adhesion coefficients, without severe slippage of the drive wheels, maintaining smooth longitudinal movement, and timely correction of the vehicle's yaw attitude. This improves the safety and reliability of obstacle avoidance operations, ensures the continuity and quality of agricultural operations, and also rationally distributes driving force, reduces power loss, and maintains overall work efficiency. Attached Figure Description

[0039] Figure 1 This is a schematic diagram of the smooth transmission transition control system for obstacle avoidance in unmanned agricultural machinery according to an embodiment of the present invention;

[0040] Figure 2 This is a flowchart of the transmission smooth transition control method for obstacle avoidance of unmanned agricultural machinery according to an embodiment of the present invention;

[0041] Figure 3 This is a flowchart illustrating a method for real-time updating of the current ground adhesion coefficient according to an embodiment of the present invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be described in detail, clearly, and completely below with reference to the accompanying drawings. It should be particularly noted that the specific embodiments described below are only for better illustrating and explaining the technical solutions of the present invention, and are intended to enable those skilled in the art to better understand and implement the present invention, and should not be construed as limiting the scope of protection of the present invention. Without departing from the spirit and substance of the present invention, those skilled in the art can modify, adjust, or make equivalent substitutions based on the content disclosed in the present invention, and these should all be considered within the scope of protection of the present invention.

[0043] Example 1:

[0044] Please see Figure 1 As shown, this embodiment discloses a transmission smooth transition control system for obstacle avoidance in unmanned agricultural machinery, including a data calculation module, a deviation calculation module, an instruction calculation module, a first instruction correction module, a second instruction correction module, and an instruction drive module. Each module is connected by wires and / or wirelessly to realize data transmission.

[0045] The data calculation module calculates the current slip ratio of each drive wheel based on real-time collected vehicle status data and updates the current ground adhesion coefficient in real time.

[0046] Vehicle status data includes vehicle position, heading angle, yaw rate, longitudinal acceleration, lateral acceleration, drive wheel speed, drive torque, and steering angle.

[0047] The vehicle position is the real-time latitude and longitude coordinates of the agricultural machinery obtained through high-precision GNSS, used to locate the vehicle's position in the farmland. The GNSS receiving antenna is usually mounted on the top of the vehicle to ensure a good field of view. The receiver outputs position data and speed information, with a typical sampling frequency of approximately 5–20 Hz. The heading angle is the absolute orientation angle of the vehicle's forward direction, which can be obtained from dual-antenna GNSS or combined inertial navigation. Typically, dual antennas are mounted on the top of the vehicle to directly calculate the heading, or it can be estimated using IMU gyroscope output combined with GNSS position changes through Kalman filtering. The sampling frequency is around 10 Hz, and can be interpolated to 100 Hz by the inertial measurement unit (IMU), providing the vehicle's current orientation for calculating heading deviation. The yaw rate is the angular velocity of the vehicle's vertical axis rotation provided by the IMU. The IMU is installed near the vehicle's center of gravity, such as at the center of the rear axle, to measure the actual vehicle body rotation. The sampling frequency is typically above 100 Hz. The yaw rate is used to monitor the vehicle's rotational dynamics, detecting changes in yaw deviation during sharp turns and skidding, for stability control. Longitudinal acceleration is the acceleration in the vehicle's forward direction measured by the IMU. Longitudinal acceleration is used to evaluate driving performance, such as comparing the expected acceleration with the actual acceleration to help identify drive slippage and adhesion. Lateral acceleration is the lateral acceleration of the vehicle measured by the IMU accelerometer. It is used to monitor the vehicle's lateral dynamic state and determine whether the vehicle is showing signs of sideslip during sharp turns. Drive wheel speed is the rotational speed of each drive wheel, measured by wheel speed sensors such as Hall sensors or encoders mounted on the wheel hub or drive shaft. Typically mounted on the rear drive wheel hubs, one set for each drive wheel, the sampling frequency can reach 100Hz or even higher, depending on the wheel speed pulses. Drive wheel speed is used to calculate tire slip ratio, and by comparing the difference between left and right wheel speeds, the degree of unilateral slippage can also be determined. Drive torque is the torque information output by the engine or motor to the drive wheels. It can be obtained from the torque estimate provided by the engine ECU or throttle opening. For electric drives, it is provided by motor current or torque sensors, installed in the powertrain system, such as in the engine crankshaft, drive shaft, or motor controller, with a sampling frequency typically between 10 and 50 Hz. Drive torque data is used, combined with wheel speed and vehicle acceleration, to estimate whether the current traction exceeds the ground grip capacity, in order to determine the tendency to slip. The steering angle is the front wheel steering angle, measured by angle sensors on the steering mechanism, such as angle encoders mounted on the steering knuckle arms or steering cylinders, with a sampling frequency of approximately 50Hz. The steering angle is used to understand the vehicle's current steering state, both for path tracking control and to determine vehicle response in conjunction with yaw rate, for example, abnormal yaw response in cases of understeer or oversteer.

[0048] The system reads vehicle status data in real time and performs necessary filtering and fusion. First, Kalman filtering is performed on GNSS positioning and IMU inertial data to calculate the vehicle's precise position coordinates and heading angle. Wheel speedometer pulses are then read and converted to calculate the real-time rotational speed of each wheel. The raw accelerometer output is denoised to obtain the vehicle's current longitudinal and lateral accelerations. All data is synchronized to the same timestamp and stored.

[0049] The method for calculating the tire slip ratio of each drive wheel includes: the tire slip ratio of each drive wheel is equal to the linear velocity corresponding to the wheel speed minus the actual vehicle speed, divided by the actual vehicle speed. If the actual vehicle speed is close to zero, a small threshold is used for approximation. The specific value of the small threshold is determined according to the agricultural machinery model and the operating scenario. Specifically, the small threshold for wheeled agricultural machinery is 0.1 m / s, and the small threshold for tracked agricultural machinery is 0.05 m / s. The initial reference value of the small threshold is determined through bench calibration tests. During field operations, the controller adaptively corrects it according to the real-time operating scenario. In dry soil operating scenarios, the initial reference value is maintained, while in wet mud operating scenarios, the small threshold is lowered by 10% to adapt to the low-adhesion, low-speed operating state, ensuring the accuracy of the slip ratio calculation near zero speed. After calculating the slip ratio of the left and right drive wheels separately, the absolute difference between the slip ratios of the left and right drive wheels is calculated and compared with a preset difference threshold. If the absolute difference between the slip ratios of the left and right drive wheels exceeds the preset difference threshold, it is determined that there is a unilateral slippage difference between the left and right drive wheels.

[0050] Please see Figure 3 As shown, the ground adhesion coefficient is calculated in real time by the vehicle's own estimation algorithm, with an update frequency consistent with the control cycle. The ground adhesion coefficient is used to determine the maximum driving force and braking force that the current ground can provide, providing control decisions to limit driving commands and avoid excessive slippage. The method for real-time updating of the ground adhesion coefficient includes: based on the vehicle dynamics model, converting the engine output torque into theoretical acceleration and comparing it with the longitudinal acceleration measured by the IMU. If the acceleration expected to be generated by the engine output torque is greater than the longitudinal acceleration and an increase in the slip ratio of the drive wheels is detected, it indicates that the driving force exceeds the ground adhesion capacity, and it is inferred that the current ground adhesion coefficient is low. When the slip ratio of the drive wheels rapidly exceeds a set threshold and the vehicle acceleration no longer increases, it is determined that the vehicle has entered a slippage state, and the estimated value of the ground adhesion coefficient is adjusted accordingly.

[0051] The system continuously monitors the difference in slip ratio between the left and right wheels. If the slip ratio of one drive wheel is significantly higher than that of the other, it indicates severe unilateral slippage. In this case, the system first obtains ground condition information from the terrain perception module on the vehicle, and simultaneously retrieves the duration of slippage of the drive wheel on that side and the corresponding change in drive torque. Then, it combines preset local working condition judgment thresholds and logic to complete the working condition judgment. When the terrain perception module shows that only the ground firmness of the area corresponding to the drive wheel on that side is lower than the preset firmness threshold and the soil moisture is higher than the preset moisture threshold, while the ground condition of the other areas meets the normal operating standards, and the duration of slippage of the drive wheel on that side is less than the preset local slippage duration threshold and the corresponding change in drive torque is less than the preset torque change threshold, it is judged as a local working condition. If it is judged as a local working condition, the current overall ground adhesion coefficient is used as the benchmark. The adhesion coefficient corresponding to the high slip ratio drive wheel on that side is weighted and corrected. The weighting coefficient is determined based on the duration of slippage. If the slippage duration is shorter, the weighting ratio of the adhesion coefficient on that side is lower. When the terrain perception module shows that the ground condition of the corresponding area of ​​the whole vehicle is low adhesion, or the slippage duration of the drive wheel on that side is greater than or equal to the preset local slippage duration threshold, or the corresponding change in drive torque is greater than or equal to the preset torque change threshold, it is determined to be a non-local working condition, that is, a working condition in which the overall ground adhesion capability is decreasing. If it is determined to be a non-local working condition, the adhesion coefficient corresponding to the slip ratio of the drive wheel on that side is used as the estimated value of the current overall ground adhesion coefficient. This ensures driving safety in local low adhesion areas and avoids waste of agricultural machinery power and reduction in operating efficiency caused by excessive restriction of overall driving force due to local working conditions.

[0052] If the slip ratio of one drive wheel is significantly higher than that of the other, but this is not considered a localized working condition, the slippage state of the drive wheel on that side can still reflect the minimum adhesion that the ground can currently provide. This avoids a mismatch between the overall control strategy and the actual ground adhesion due to poor adhesion conditions on one side. By comprehensively considering the changes in drive wheel slip ratio, the deviation between theoretical acceleration and actual longitudinal acceleration, and the correction results for the difference in slip ratio between the left and right wheels, the ground adhesion coefficient is updated in real time, dynamically increasing or decreasing the ground adhesion coefficient according to the actual working conditions.

[0053] The deviation calculation module calculates lateral deviation and heading deviation based on the current vehicle position and heading angle, as well as the pre-loaded target path.

[0054] The target path is a pre-provided desired driving trajectory, a series of coordinate points or path equations, representing a safe route around obstacles. The target path is obtained through offline or real-time path planning and stored in the controller's memory, read as needed during the control cycle, and the interval between adjacent points can be 1 meter or less.

[0055] Based on the current vehicle position, the nearest target point or tangent segment on the target path is found, and the lateral deviation is calculated. Specifically, the vehicle position is projected onto the target path, and the vertical offset distance is calculated. A positive vertical offset distance indicates the vehicle is deviating to the right of the path, and a negative vertical offset distance indicates the vehicle is deviating to the left of the path. This vertical offset distance is used as the lateral deviation. The heading deviation is calculated by taking the difference between the vehicle's current heading angle and the tangent direction of the target path. For example, an angle difference within the range of [-180°, 180°] can be used to represent the heading deviation. For smoothing, a low-pass filter is applied to the continuous periodic deviation signal to avoid control jitter caused by sensor noise. Lateral and heading deviations quantitatively describe the vehicle's position and orientation away from the desired trajectory, providing feedback for control decisions.

[0056] The command calculation module calculates preliminary control commands based on lateral deviation, heading deviation, vehicle status data, and environmental information.

[0057] The steering controller is designed based on lateral deviation and heading deviation, using a PID or pure tracking algorithm. The steering angle command is set to K1 multiplied by the heading deviation plus K2 multiplied by the lateral deviation, where K1 is the proportional control gain corresponding to the heading deviation and K2 is the proportional control gain corresponding to the lateral deviation. The values ​​of K1 and K2 need to be determined in conjunction with the vehicle's current speed, the curvature of the target path, and the real-time ground adhesion coefficient. The value of K1 decreases as the vehicle speed increases and increases as the curvature of the target path increases to avoid excessive heading deviation correction at high speeds, which could cause vehicle sway, and to ensure rapid correction of heading deviation on paths with high curvature. The value of K2 increases as the absolute value of the lateral deviation increases and decreases as the ground adhesion coefficient decreases to ensure sufficient correction strength when the lateral deviation is large and to prevent sideslip caused by excessive lateral correction under low adhesion coefficient conditions. The initial reference values ​​of K1 and K2 are determined through bench calibration tests. During vehicle operation, the controller adjusts the steering angle based on real-time vehicle status data and environmental information. The fuzzy adaptive control algorithm performs adaptive fine-tuning. When the change in the vehicle's current speed compared to the previous control cycle exceeds a preset speed change threshold, or the change in the target path curvature compared to the previous control cycle exceeds a preset curvature change threshold, or the change in the real-time ground adhesion coefficient compared to the previous control cycle exceeds a preset adhesion coefficient change threshold, the adaptive fine-tuning process is triggered. The preset speed change threshold is 5 km / h, the preset curvature change threshold is 0.01 rad / m, and the preset adhesion coefficient change threshold is 0.1. The fuzzy adaptive control algorithm takes the vehicle speed deviation, target path curvature deviation, and ground adhesion coefficient deviation as inputs and the gain correction values ​​of K1 and K2 as outputs. It outputs the corresponding gain correction values ​​through a preset fuzzy rule base to dynamically adjust the current values ​​of K1 and K2. The adjusted K1 and K2 must meet the preset upper and lower gain limits to ensure the stability of steering control. This allows for the practical verification of the strategy and its adaptation to different operating conditions. The steering control algorithm incorporates look-ahead distance considerations. This means that when calculating the steering angle command, a path point located at a preset distance ahead of the vehicle's current position on the target path is selected as a preview point. The steering control logic is adjusted based on the path tangent direction and lateral offset corresponding to the preview point. This look-ahead distance consideration compensates for the response delay of the steering actuator, preventing lag in steering adjustments due to relying solely on the deviation of the current position, thus enabling the vehicle to more smoothly align with the target path. The steering control algorithm, combining lateral deviation, heading deviation, preset proportional control gain, and look-ahead distance, generates the required steering correction. The steering angle command is the sum of the vehicle's current steering angle and the steering correction.

[0058] The method for determining longitudinal control commands includes: if the obstacle distance is less than a set safe distance threshold, the target speed is adjusted to a value lower than the current agricultural operation speed, achieved by reducing drive torque output or applying light braking; when the obstacle distance is less than the safe distance threshold, the required deceleration is calculated based on the vehicle's current speed and the relative position of the obstacle, enabling the vehicle to decelerate to a preset safe speed or complete obstacle avoidance maneuvers within a safe area before reaching the obstacle. When the vehicle travels to a distance greater than or equal to the safe distance threshold and has entered the preset obstacle avoidance completion area, the target speed is gradually increased back to the original agricultural operation speed to ensure the overall efficiency of agricultural operations. Obstacle distance refers to the distance information between the vehicle and the obstacle ahead. This information can be obtained by acquiring environmental point clouds using a LiDAR installed on the front or top of the vehicle, and then detecting the nearest obstacle distance. The LiDAR has horizontal full-circle scanning capability, outputs detection data at a fixed frequency, and its detection range covers a preset range of tens of meters. It can be supplemented with cameras or millimeter-wave radar to detect obstacles and improve the comprehensiveness of environmental perception. The longitudinal control commands aim to maintain a smooth speed change, using acceleration / deceleration commands from the cruise speed setting or position loop output. However, tire slippage limitations are not considered. If the current path has a large curvature or significant heading deviation, the target speed is reduced accordingly to prevent sideslip. The target speed value is directly adjusted, and by matching corresponding acceleration / deceleration commands, the vehicle's actual speed gradually approaches the reduced target speed. This reduces the risk of sideslip caused by high-speed passage through paths with large curvature or correction of large heading deviations. The final preliminary control commands consist of longitudinal and lateral control commands, with the lateral control command being the steering angle command.

[0059] The first command correction module corrects the longitudinal control command in the initial control command based on the current slip ratio and current ground adhesion coefficient of each drive wheel, thus obtaining the corrected longitudinal control command.

[0060] Methods for correcting longitudinal control commands include: determining the maximum permissible acceleration based on the current ground adhesion coefficient, and controlling the drive wheel slip ratio within a set reasonable range to achieve optimal adhesion coefficient utilization. For example, if the acceleration or deceleration required by the initial longitudinal control command would cause the drive wheel slip ratio to exceed a set safety threshold, then the drive torque output is reduced, the throttle opening is decreased, or appropriate braking is performed to bring the drive wheel slip ratio back to a safe range. Specifically, the relationship between the target acceleration and μ×g is compared. μ×g is the product of the current ground adhesion coefficient and gravitational acceleration. This value represents the maximum longitudinal acceleration that the current ground can provide to the vehicle. Physically, the ground adhesion coefficient determines the maximum driving force that the ground can transmit to the vehicle. Combining vehicle dynamics and Newton's second law, the maximum driving force can be converted into the maximum longitudinal acceleration that the vehicle can obtain, i.e., μ×g. Therefore, only by controlling the target acceleration within the μ×g range can it be ensured that the driving force does not exceed the ground adhesion limit, avoiding excessive tire slippage. When the acceleration required by the initial longitudinal control command exceeds μ×g, the acceleration command is reduced to the μ×g level to prevent tire slippage. Similarly, during deceleration and braking, if drive wheel lock-up is detected, the pressure supply from the master cylinder to the corresponding drive wheel brake caliper is reduced, and a closed-loop pressure feedback adjustment is initiated. Based on real-time wheel speed data from the drive wheel speed sensors, the opening of the brake pressure regulating valve is adjusted to gradually reduce the hydraulic pressure in the brake calipers until the drive wheel speed returns to a reasonable range matching the vehicle's actual speed. At this point, the drive wheel lock-up is released, preventing the vehicle from skidding. Throughout this process, the drive wheel slip ratio is continuously monitored, and the throttle is adjusted using closed-loop control to maintain the actual drive wheel slip ratio near its optimal value.

[0061] The revised longitudinal control commands include the actual throttle opening or braking force, which, while meeting the original deceleration and acceleration requirements, will not cause excessive slippage of the drive wheels, ensuring smooth and controlled longitudinal movement of the vehicle and providing stable traction.

[0062] The second instruction correction module corrects the lateral control instruction in the initial control instruction based on vehicle status data, thus obtaining the corrected lateral control instruction.

[0063] When performing steering obstacle avoidance, the vehicle's yaw motion is monitored to prevent yaw loss of control due to slippage. Preset yaw rate thresholds and slip rate difference thresholds are determined through bench dynamics calibration tests combined with multi-condition field tests. First, the yaw response and drive wheel slippage state of agricultural machinery under different adhesion coefficients and steering commands are simulated in a bench environment to obtain initial baseline values ​​for the corresponding thresholds. Then, field tests are conducted in typical farmland operation scenarios. The initial baseline values ​​are corrected based on the actual yaw stability effect and slippage detection accuracy of the agricultural machinery. Different threshold baselines are set for different agricultural machinery models. For example, the initial baseline value for the yaw rate threshold for wheeled agricultural machinery is 0.05 rad / (s·s), and the initial baseline value for the yaw rate threshold for tracked agricultural machinery is 0.03 rad / (s·s). The yaw rate threshold for tracked agricultural machinery is lower than that for wheeled agricultural machinery to adapt to the slower yaw response of tracked agricultural machinery. The slip rate threshold for wheeled agricultural machinery is lower than that for wheeled agricultural machinery. The initial baseline value for the slip rate difference threshold is 8%, while the initial baseline value for the slip rate difference threshold for tracked agricultural machinery is 5%. The slip rate difference threshold for tracked agricultural machinery is lower than that for wheeled agricultural machinery, which is suitable for the characteristic that unilateral slippage of tracked agricultural machinery is more likely to cause overall yaw. During vehicle operation, the controller will dynamically adjust the threshold according to the real-time ground adhesion coefficient. For example, when the ground adhesion coefficient decreases, the yaw rate change threshold is lowered by 10% and the slip rate difference threshold is lowered by 15% to improve the control sensitivity under low adhesion conditions. When the yaw rate change exceeds the calibrated and adjusted yaw rate change threshold, it is determined that a sharp yaw change has occurred. Or, when the absolute difference in slip rates between the left and right drive wheels exceeds the calibrated and adjusted slip rate difference threshold, it is determined that excessive unilateral wheel slippage has occurred, which leads to instantaneous vehicle yaw and triggers yaw stabilization control.

[0064] If the actual yaw rate exceeds the expected yaw rate threshold calculated based on the current steering command, it indicates oversteer. The engine output torque is then reduced to a preset safe torque range. The absolute difference in slip ratio between the left and right drive wheels is calculated and compared to a preset slip ratio difference threshold. If the difference exceeds the preset threshold, the side with the higher slip ratio is identified as the severely slipping side. The initial benchmark value of the preset slip ratio difference threshold is determined by bench dynamics calibration tests combined with field multi-condition tests: 8% for wheeled agricultural machinery and 5% for tracked agricultural machinery. This threshold is dynamically adjusted based on the current ground adhesion coefficient. When the current ground adhesion coefficient decreases, the preset slip ratio difference threshold is lowered by 15% to improve the sensitivity under low-adhesion conditions, thus avoiding misjudgment of the braking distribution target during the oversteer correction phase. The hydraulic pressure in the brake caliper corresponding to the severely slipping drive wheel is increased, providing that drive wheel with a preset braking torque, and a single-sided braking operation is performed on the severely slipping drive wheel. By applying braking torque to the slipping drive wheel, the free spin of that drive wheel is suppressed. The power distribution module adjusts the traction distribution ratio between the two drive wheels, allocating more traction to the other drive wheel, thus generating a corrective yaw moment to counteract excessive vehicle yaw. This operation is equivalent to introducing differential torque-assisted steering. When conventional steering control cannot meet the attitude correction requirements, the difference in acceleration and deceleration between the two drive wheels generates additional steering torque, achieving precise correction of the vehicle's attitude.

[0065] When the right drive wheel spins freely, causing the vehicle to yaw to the left, the braking duration applied to the right drive wheel is not a fixed value. Its initial reference duration is determined through bench calibration tests and is dynamically adjusted based on the vehicle's current speed, the current ground adhesion coefficient, and the degree of drive wheel slippage. The higher the vehicle's current speed, the longer the braking duration; the lower the current ground adhesion coefficient, the shorter the braking duration; and the more severe the drive wheel slippage (i.e., the higher the drive wheel slip ratio), the longer the braking duration. Upper and lower limits for the braking duration are set to avoid control risks caused by excessively long or short durations. For example, the upper limit for the braking duration is set to 500ms, and the lower limit to 1ms. 00ms is used to adapt to the correction requirements under different working conditions, prevent insufficient correction under high-speed and high-adhesion conditions, and avoid violent fluctuations in vehicle attitude under low-speed and low-adhesion conditions. The braking duration of the right drive wheel is dynamically adjusted to achieve deceleration, while maintaining the current traction output of the left drive wheel, forcing the vehicle to correct its yaw direction towards the preset path. When the left drive wheel spins and causes the vehicle to yaw to the right, the braking duration of the left drive wheel is determined by the same dynamic adjustment logic to achieve deceleration, while maintaining the current traction output of the right drive wheel, forcing the vehicle to correct its yaw direction towards the preset path.

[0066] When a vehicle is detected to have deviated from its path due to slippage, the initial steering angle command can be temporarily increased within a preset steering angle compensation range based on the real-time calculated lateral and directional deviations. This increase in steering angle command must not exceed a preset compensation threshold until the vehicle's actual yaw state returns to a preset safe range, prioritizing yaw stability. This process employs high-speed closed-loop control, continuously comparing the real-time yaw deviation rate of change with a preset yaw stability threshold to determine the specific force and duration of differential braking until the vehicle's actual yaw state returns to a preset safe range.

[0067] The revised lateral control commands include final steering angle command, differential braking, and torque command. These revised lateral control commands ensure that even if drive slippage occurs during obstacle avoidance maneuvers, the vehicle's lateral attitude can be corrected promptly, preventing yaw deviation from accumulating and causing severe trajectory deviation.

[0068] The command-driven module is used to drive the vehicle by taking the modified longitudinal control command and the modified lateral control command as the final control command.

[0069] The final command is sent to the vehicle's actuators. The steering actuator adjusts the front wheel steering according to the target steering angle. If a steering angle position sensor is equipped, a closed loop ensures that the command value is achieved. The drive system outputs corresponding torque according to throttle or braking commands; for example, the engine control unit adjusts the throttle opening, and the hydraulic brake actuator applies differential braking force. During execution, the controller continuously monitors sensor feedback, including whether the actual steering angle is accurately tracked, whether the drive motor or engine response reaches the target torque, and changes in vehicle acceleration and yaw rate. If there is a deviation between the actual execution and the command, it is corrected by adjusting the control input in the next control cycle, forming a closed-loop control.

[0070] The vehicle begins to navigate around the obstacle according to the corrected steering angle, while its longitudinal speed changes smoothly without severe wheel slippage. With the help of sensor feedback, the system enters the next control cycle, continuously repeating the above steps to dynamically correct the vehicle's attitude, ultimately guiding it to bypass the obstacle along the planned path and resume the predetermined route. Throughout the process, all modules work in coordination, avoiding the sudden slippage of the drive wheels and the resulting yaw and trajectory deviations that might occur with only longitudinal smooth control, ensuring the stability and accuracy of the unmanned agricultural machinery's obstacle avoidance maneuvering in complex terrain.

[0071] Example 2:

[0072] Please see Figure 2 As shown, this embodiment provides a transmission smooth transition control method for obstacle avoidance in unmanned agricultural machinery, including:

[0073] Based on real-time collected vehicle status data, the current slip ratio of each drive wheel is calculated, and the current ground adhesion coefficient is updated in real time.

[0074] Based on the current vehicle position and heading angle, as well as the pre-loaded target path, calculate the lateral deviation and heading deviation;

[0075] Based on lateral deviation, heading deviation, vehicle status data, and environmental information, preliminary control commands are calculated.

[0076] Based on the current slip ratio and current ground adhesion coefficient of each drive wheel, the longitudinal control command in the initial control command is corrected to obtain the corrected longitudinal control command.

[0077] Based on vehicle status data, the lateral control command in the initial control command is corrected to obtain the corrected lateral control command.

[0078] The modified longitudinal control command and the modified lateral control command are used as the final control commands to drive the vehicle.

[0079] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0080] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A transmission smooth transition control method for obstacle avoidance in unmanned agricultural machinery, characterized in that, include: Based on real-time collected vehicle status data, the current slip ratio of each drive wheel is calculated, and the current ground adhesion coefficient is updated in real time. Based on the current vehicle position and heading angle, as well as the pre-loaded target path, calculate the lateral deviation and heading deviation; Based on lateral deviation, heading deviation, vehicle status data, and environmental information, preliminary control commands are calculated. Based on the current slip ratio and current ground adhesion coefficient of each drive wheel, the longitudinal control command in the initial control command is corrected to obtain the corrected longitudinal control command. Based on vehicle status data, the lateral control command in the initial control command is corrected to obtain the corrected lateral control command. The modified longitudinal control command and the modified lateral control command are used as the final control commands to drive the vehicle.

2. The transmission smooth transition control method for obstacle avoidance in unmanned agricultural machinery according to claim 1, characterized in that, The current slip ratio of each drive wheel is calculated by comparing the linear velocity corresponding to the wheel speed of the drive wheel with the actual speed of the vehicle. Based on the current slip ratio of the drive wheels on the left and right sides of the vehicle and the preset difference threshold, it is determined whether there is a unilateral slip difference between the drive wheels on the left and right sides.

3. The transmission smooth transition control method for obstacle avoidance in unmanned agricultural machinery according to claim 2, characterized in that, The engine output torque is converted into theoretical acceleration and compared with the actual longitudinal acceleration measured by the IMU to obtain the current ground adhesion coefficient.

4. The transmission smooth transition control method for obstacle avoidance in unmanned agricultural machinery according to claim 3, characterized in that, If there is a difference in slippage on one side of the drive wheels on the left and right sides, then determine whether the slippage on one side is a local working condition. If it is determined to be a local working condition, the current overall ground adhesion coefficient is used as the benchmark, and the adhesion coefficient corresponding to the side with the higher slip ratio value of the left and right drive wheels is weighted and corrected. The weighting coefficient is determined according to the duration of slippage. If the condition is determined to be non-local, the adhesion coefficient corresponding to the slip ratio of the drive wheel with the higher slip ratio value on the left and right sides shall be used as the current ground adhesion coefficient.

5. The transmission smooth transition control method for obstacle avoidance in unmanned agricultural machinery according to claim 1, characterized in that, The method for correcting the longitudinal control command in the initial control command is to determine the maximum allowable acceleration based on the current ground adhesion coefficient.

6. The transmission smooth transition control method for obstacle avoidance in unmanned agricultural machinery according to claim 5, characterized in that, Methods for determining the maximum permissible acceleration based on the current ground adhesion coefficient include: The maximum longitudinal acceleration is calculated based on the current ground adhesion coefficient and gravitational acceleration. If the target acceleration exceeds the maximum longitudinal acceleration, the target acceleration will be reduced to the maximum longitudinal acceleration.

7. The transmission smooth transition control method for obstacle avoidance in unmanned agricultural machinery according to claim 6, characterized in that, During the braking process, if a lock-up of the drive wheel is detected, the pressure supply from the master cylinder to the corresponding drive wheel brake cylinder is reduced, and the opening of the brake pressure regulating valve is adjusted according to the drive wheel speed to reduce the hydraulic pressure value in the brake brake cylinder until the drive wheel speed matches the actual vehicle speed, thus releasing the lock-up of the drive wheel.

8. The transmission smooth transition control method for obstacle avoidance in unmanned agricultural machinery according to claim 1, characterized in that, When the yaw rate change exceeds the preset yaw rate change threshold, or the absolute difference between the slip rates of the left and right drive wheels exceeds the preset slip rate difference threshold, the lateral control command in the initial control command is corrected.

9. The transmission smooth transition control method for obstacle avoidance in unmanned agricultural machinery according to claim 8, characterized in that, If the actual yaw rate exceeds the expected yaw rate threshold calculated based on the current steering command, the engine torque output will be reduced to the preset safe torque range. Calculate the absolute difference between the slip ratios of the left and right drive wheels and compare it with a preset slip ratio difference threshold. If the difference exceeds the preset slip ratio difference threshold, the side with the higher slip ratio value of the left and right drive wheels is determined to be the side with severe slippage. Apply single-sided braking to the drive wheel on the side with the most severe slippage; When one drive wheel spins freely and causes the vehicle to yaw in the opposite direction, brakes are applied to that drive wheel to slow it down and correct the vehicle's yaw direction toward the target path. When a vehicle is detected to have deviated from its path due to slippage, the steering angle command is increased to compensate for the path deviation, and the increase in the steering angle command does not exceed a preset compensation threshold.

10. A transmission smooth transition control system for obstacle avoidance in unmanned agricultural machinery, used to implement the transmission smooth transition control method for obstacle avoidance in unmanned agricultural machinery as described in any one of claims 1-9, characterized in that, include: The data calculation module calculates the current slip ratio of each drive wheel based on real-time collected vehicle status data and updates the current ground adhesion coefficient in real time. The deviation calculation module calculates the lateral deviation and heading deviation based on the current vehicle position and heading angle, as well as the pre-loaded target path. The command calculation module calculates preliminary control commands based on lateral deviation, heading deviation, vehicle status data, and environmental information. The first command correction module corrects the longitudinal control command in the initial control command based on the current slip ratio and the current ground adhesion coefficient of each drive wheel, and obtains the corrected longitudinal control command. The second instruction correction module corrects the lateral control instruction in the initial control instruction based on vehicle status data, and obtains the corrected lateral control instruction. The command-driven module is used to drive the vehicle by taking the modified longitudinal control command and the modified lateral control command as the final control command.

Images