Self-adaptive coordination control method and device for snow shovels of snow pressing vehicle

Abstract

The invention provides a self-adaptive coordination control method and device for a snow shovel of a snow pressing vehicle, and relates to the technical field of snow pressing vehicles. The method comprises the following steps: establishing a kinematic model of the snow shovel, and determining a mapping relation between a shovel tip pose and motion parameters of each driving joint; acquiring topographic information of a snowroad in front of the snowmobile in real time, wherein the topographic information comprises gradient and distance information; determining a target attitude of the snow shovel according to the gradient in the topographic information, and resolving based on the kinematic model to obtain a target control quantity of a driving system; and based on a self-adaptive control strategy, carrying out coordination control on a driving system of the snow shovel according to the target control quantity, so that the actual posture of the snow shovel tracks the target posture to adapt to the change of a snow track. According to the method, the posture of the snow shovel can be changed along with the slope of the snow track, the quality and efficiency of snow track leveling operation are remarkably improved, the working intensity of operators is reduced, and the adaptability and the operation stability of the snow pressing vehicle under complex and changeable snow conditions are improved.

Description

Technical Field

[0001] This invention relates to the field of snow groomer technology, and in particular to an adaptive coordination control method and device for snow groomer snowplows. Background Technology

[0002] Snow groomers are key equipment for ski resort maintenance, and their snowplowing performance directly affects the smoothness of the ski slopes and the skiing experience. Traditional snow groomer operations mainly rely on the operator's experience to manually adjust the snowplow's posture. However, due to the large size of the snowplow, the wide range of motion, and the long operating distance, manual control is difficult to respond accurately to changes in the slope in real time. This often results in snowplowing being too deep and jamming, or snowplowing being too shallow and failing to effectively remove the snow layer, affecting operational efficiency and ski slope quality.

[0003] To address the aforementioned issues, existing technologies have attempted to introduce automated control methods, such as using sensors to detect ski slope height and employing preset control programs to adjust the stroke of hydraulic cylinders.

[0004] However, in existing snow groomer operations, snow shovels cannot accurately track changes in the snow track terrain in real time, affecting the accuracy of snow pushing depth control and the smoothness of the snow track. Summary of the Invention

[0005] To address the technical problem in existing technologies where snowplows struggle to accurately track changes in snow terrain in real time, thus affecting the accuracy of snowplow depth control and snow track smoothness, this invention provides an adaptive coordinated control method and device for snowplows used in snow groomers. The technical solution is as follows:

[0006] On the one hand, an adaptive coordinated control method for snow groomer snowplows is provided, including the following steps:

[0007] Establish a kinematic model of the snow shovel and determine the mapping relationship between the shovel tip posture and the motion parameters of each drive joint;

[0008] Real-time acquisition of terrain information of the snow track in front of the snow groomer, including slope and distance information;

[0009] Based on the slope in the terrain information, the target posture of the snow shovel is determined, and the target control quantity of the drive system is calculated based on the kinematic model.

[0010] Based on an adaptive control strategy, the drive system of the snowplow is coordinated and controlled according to the target control quantity, so that the actual posture of the snowplow tracks the target posture to adapt to changes in the snow track.

[0011] Optionally, establishing the kinematic model includes:

[0012] Based on the theory of multibody kinematics, a coordinate system is established at the key hinge points of the kinematic chain of the snow shovel mechanism.

[0013] By using homogeneous coordinate transformation, the functional expression of the shovel tip pose with respect to the rotation angles of each joint is derived.

[0014] Optionally, acquiring the terrain information in real time includes:

[0015] Acquire point cloud data generated by lidar scanning the ski slope ahead;

[0016] The point cloud data is organized, and ground points and non-ground points are segmented according to dynamically adjusted discrimination thresholds;

[0017] Based on the segmented ground point cloud data, the slope and distance of the ski slope ahead are calculated.

[0018] Optionally, the dynamically adjusted discrimination threshold is determined based on the geometric installation parameters of the lidar, the angle information of the scanning beam, and the real-time estimated terrain slope information.

[0019] Optionally, calculating the slope of the ski run ahead includes:

[0020] Select multiple adjacent scan line data from the ground point cloud;

[0021] Calculate the slope information representing topographic changes in each scan line data;

[0022] The slope information calculated from multiple scan lines is fused and processed to output the slope value.

[0023] Optionally, the target control quantity of the drive system is calculated based on the kinematic model, including:

[0024] Based on the target pose of the shovel tip, the required rotation angle of each joint is calculated through inverse kinematics.

[0025] Based on the geometric relationship between the joint rotation angle and the extension / retraction of the drive mechanism, the target extension / retraction of each drive mechanism is obtained.

[0026] Optionally, the adaptive control strategy includes:

[0027] A reference model is pre-defined to characterize the desired dynamic performance;

[0028] Control is achieved using a parameter adaptive regulation law;

[0029] By comparing the error between the output of the controlled system and the output of the reference model, the control parameters are dynamically adjusted so that the controlled system tracks the dynamic characteristics of the reference model.

[0030] Optionally, the adaptive adjustment law for the parameters includes:

[0031] Construct a filter auxiliary signal vector containing the system input and output signals;

[0032] The parameter update law is designed based on stability theory, and the control parameters are adjusted in real time according to the tracking error and the auxiliary signal vector.

[0033] Optionally, the method enables the snowplow to automatically adjust the tip posture under various working conditions where the relative posture of the vehicle body and the snow track changes, so as to maintain a predetermined relative relationship between the tip and the snow track surface.

[0034] On the other hand, an adaptive coordination control device for snow groomer snowplows is provided, the device comprising:

[0035] The model building module is used to build the kinematic model of the snow shovel and determine the mapping relationship between the shovel tip posture and the motion parameters of each drive joint;

[0036] The terrain acquisition module is used to acquire real-time terrain information of the snow track in front of the snow groomer, including slope and distance information;

[0037] The attitude determination module is used to determine the target attitude of the snow shovel based on the slope in the terrain information, and to calculate the target control quantity of the drive system based on the kinematic model.

[0038] The control module is used to coordinate and control the drive system of the snowplow based on the target control quantity according to the adaptive control strategy, so that the actual posture of the snowplow tracks the target posture to adapt to changes in the snow track.

[0039] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0040] The method provided in this embodiment integrates high-precision kinematic modeling, real-time terrain perception based on lidar, and robust adaptive control technology to achieve intelligent and precise snow grooming operations. It effectively solves problems such as uneven snow removal depth, snow jamming, or missed areas caused by long visibility and judgment delays in traditional manual operations. It can automatically and quickly adjust the snow plow's posture to follow changes in the slope of the snow track, significantly improving the quality and efficiency of snow track leveling operations, reducing operator workload, and enhancing the adaptability and operational stability of snow groomers in complex and variable snow conditions. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1This is a flowchart of an adaptive coordinated control method for snow groomers and snowplows provided in an embodiment of the present invention;

[0043] Figure 2 A simplified coordinate diagram of a snowplow for a snow groomer provided in an embodiment of the present invention; Figure 3 A schematic diagram for slope calculation provided in an embodiment of the present invention;

[0044] Figure 4 A flowchart of snow track point cloud data processing is provided for an embodiment of the present invention;

[0045] Figure 5 A reference model adaptive control system diagram provided for an embodiment of the present invention;

[0046] Figure 6 A schematic diagram of a Simulink control system provided in an embodiment of the present invention;

[0047] Figure 7 A schematic diagram of a reference system Simulink structure provided in an embodiment of the present invention;

[0048] Figure 8 A time-domain characteristic curve and error diagram of um=2(t) provided for an embodiment of the present invention;

[0049] Figure 9 A schematic diagram of the time-domain characteristic curve and error of um=sin(t) provided in an embodiment of the present invention;

[0050] Figure 10 A time-domain characteristic curve and error diagram of um=3sin(5t)-2sin(t) provided for embodiments of the present invention;

[0051] Figure 11 A schematic diagram of the time-domain characteristic curve of the input being the change in the hydraulic cylinder, provided in an embodiment of the present invention;

[0052] Figure 12 A time-domain characteristic curve and error diagram of the input being the change in the forward and backward tilting cylinder, provided for an embodiment of the present invention;

[0053] Figure 13 A time-domain characteristic curve and error diagram of lifting cylinder length as input are provided for an embodiment of the present invention;

[0054] Figure 14 This is a structural block diagram of an adaptive coordination control device for snow groomers and snowplows provided in an embodiment of the present invention. Detailed Implementation

[0055] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0056] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0057] In this embodiment of the invention, sometimes a subscript such as W1 may be mistakenly written as a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0058] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0059] This invention provides an adaptive coordinated control method for snow groomer snowplows. For example... Figure 1 The flowchart shown is for the adaptive coordinated control method of snow groomer and snowplow. The processing flow of this method may include the following steps:

[0060] 100. Establish a kinematic model of the snow shovel and determine the mapping relationship between the shovel tip position and the motion parameters of each driving joint.

[0061] In this step, the snowplow mechanism is analyzed using multibody kinematics theory. The snowplow mechanism is considered a kinematic chain consisting of the frame, front connecting frame, swing frame, and snowplow body connected by hinge points. First, Cartesian coordinate systems are established at the key hinge points of this kinematic chain: the hinge between the frame and the front connecting frame, the hinge between the front connecting frame and the swing frame, and the snowplow tip. Then, using homogeneous coordinate transformation, a series of rotation and translation transformation matrices are used to establish the pose transformation relationship of the end-point snowplow tip coordinate system relative to the frame base coordinate system. This process ultimately derives precise mathematical function expressions for the spatial position and attitude angle of the snowplow tip with respect to the rotation angles of the front connecting frame and the swing frame, thus fully describing the mapping relationship between the snowplow tip pose and the motion parameters of each drive joint, providing a precise mathematical model foundation for subsequent attitude planning and control.

[0062] 200. Real-time acquisition of terrain information of the snow track in front of the snow groomer, including slope and distance information.

[0063] In this step, a multi-line lidar sensor installed on the top of the snow groomer's cab continuously scans the snow track environment in front of the vehicle to acquire raw 3D point cloud data. The point cloud data is then processed: first, irrelevant point clouds from the vehicle itself and its rear are removed; then, based on the lidar's scanning sequence, the point cloud is organized in an ordered manner according to horizontal and vertical angles.

[0064] 300. Based on the slope in the terrain information, determine the target posture of the snow shovel, and calculate the target control quantity of the drive system based on the kinematic model.

[0065] In this step, the control system's core objective is to keep the shovel tip parallel to the ski slope surface. Once the slope information ahead is acquired in real time, the target pose angle that the shovel tip needs to achieve is determined based on this slope value. Subsequently, the shovel tip target pose, including the target position coordinates and the target pose angle, is input into a pre-established kinematic model for inverse kinematics calculation. This calculation process, based on geometric relationships, inversely solves for the target angles that the front connecting frame and the swing frame need to rotate to achieve the target pose.

[0066] Furthermore, based on these target joint angles, and combined with the fixed geometric positions of the lifting cylinder and the tilting cylinder with each structural component hinge points, the target length that each hydraulic cylinder needs to extend or retract is calculated through analytical geometric relationships. This target length is the target control quantity of the drive system.

[0067] 400. Based on an adaptive control strategy, the drive system of the snowplow is coordinated and controlled according to the target control quantity, so that the actual posture of the snowplow tracks the target posture to adapt to changes in the snow track.

[0068] In this step, a reference model adaptive control strategy is employed to perform closed-loop control of the snowplow's hydraulic drive system. This strategy first constructs a reference model with ideal dynamic response characteristics, representing the system's desired output performance. The core of the controller includes an adaptive mechanism with online adjustable parameters. During control, the calculated target control quantity is used as input, while the actual extension and retraction of the hydraulic cylinder is collected in real time as feedback. The adaptive mechanism compares the error between the actual output of the controlled hydraulic system and the ideal output of the reference model, and dynamically adjusts the internal parameters of the controller according to a pre-designed parameter adaptive law based on stability theory, thereby correcting the control signal in real time. This design enables the hydraulic system to quickly, smoothly, and accurately drive the hydraulic cylinder to the target position even when facing uncertainties such as load changes and fluctuations in oil characteristics. Ultimately, this achieves accurate tracking of the snowplow's actual posture to the target posture, allowing it to adapt to constantly changing snow terrain.

[0069] The method provided in this embodiment integrates high-precision kinematic modeling, real-time terrain perception based on lidar, and robust adaptive control technology to achieve intelligent and precise snow grooming operations. It effectively solves problems such as uneven snow removal depth, snow jamming, or missed areas caused by long visibility and judgment delays in traditional manual operations. It can automatically and quickly adjust the snow plow's posture to follow changes in the slope of the snow track, significantly improving the quality and efficiency of snow track leveling operations, reducing operator workload, and enhancing the adaptability and operational stability of snow groomers in complex and variable snow conditions.

[0070] In one embodiment of the present invention, establishing the kinematic model in step 100 includes:

[0071] 110. Based on the theory of multibody kinematics, establish a coordinate system at the key hinge points of the kinematic chain of the snow shovel mechanism.

[0072] In this step, the mechanical structure of the snowplow is abstracted as a series of kinematic chains consisting of the vehicle body, front connecting frame, swing frame, and shovel body. Specifically, coordinate systems are established at three core locations of the kinematic chain: the hinge point between the vehicle body and the front connecting frame is used as the origin of the base coordinate system; the hinge point between the front connecting frame and the swing frame is used as the origin of the intermediate coordinate system; and the shovel tip is used as the origin of the end-effector coordinate system. Then, following standard robotics modeling methods, the link lengths and joint angle parameters between adjacent coordinate systems are determined.

[0073] 120. Using the homogeneous coordinate transformation method, derive the functional expression of the shovel tip pose with respect to the rotation angles of each joint.

[0074] In this step, the transformation matrices describing the pose relationship between adjacent coordinate systems are multiplied in sequence according to the kinematic chain using the homogeneous coordinate transformation method, ultimately yielding a comprehensive transformation matrix. This matrix explicitly expresses the pose of the end-effector tip coordinate system relative to the base coordinate system, and its elements constitute functional expressions of the tip position coordinates and attitude angles with respect to the front connecting frame rotation angle and the swing frame rotation angle, thus completing the construction of the forward kinematics model.

[0075] Please see Figure 2 In this diagram, point O represents the hinge point between the front connecting frame and the frame; points A and D represent the hinge points between the lifting cylinder and the frame and the front connecting frame, respectively; points B and C represent the hinge points between the tilting cylinder and the front connecting frame and the swing frame, respectively; point E represents the hinge point between the front connecting frame and the swing frame; point F represents the snowplow tip; and point OE represents the front connecting frame. Since there is no relative movement between the swing frame and the snowplow, they can be considered as a whole, represented by EF.

[0076] The DH (Denavit-Hartenberg) method follows the principles of: the i-th joint rotates around Z... i-1 The axis rotates, X i The axis must be perpendicular to Z.i-1 axis, Y i The axes are established according to the right-hand rule. The base coordinate system {O} is located at the rotation center O of the front connecting frame, the rotation center E of the snow shovel is used to establish the coordinate system {E}, and finally the coordinate system {F} is established at the tip F of the snow shovel.

[0077] Given the connecting frame parameters and the variables of each joint, the DH parameter table of the snowplow is shown in Table 1:

[0078] Table 1. DH Parameters for Snow Grooms and Snow Shovels

[0079]

[0080] Based on the coordinate system established above and the determined DH parameter table, the coordinates of the snow groomer's snowplow tip are ( Thus, the transformation matrix is ​​determined. The expression is:

[0081] Formula 1

[0082] Substituting the parameters from Table 1, the transformation matrix is ​​shown in the following equation:

[0083] Formula 2

[0084] Formula 3

[0085] Formula 4

[0086] According to the DH method, by multiplying the equations sequentially, we obtain the transformation matrix of the coordinate system {F} of the snowplow tip relative to the hinge point {O} of the frame and the front connecting frame.

[0087] Formula 5 Let the coordinates of the snowplow tip F of the snow groomer be ( ), pose angle is The coordinates of the snow groomer tip F are:

[0088] Formula 6

[0089] Formula 7

[0090] Formula 8

[0091] If the front connecting frame angle can be determined Corner of the display stand If the hinge points in the front connecting frame and the swing frame are located in the relative coordinate system {O}, {E}, {F} respectively, then the position of the snow groomer's snowplow tip F relative to the base coordinate is also determined.

[0092] In the snowplow mechanism of a snow groomer, the joint angles of the front connecting frame and the swing frame can be calculated from the known extension and retraction of the hydraulic cylinders. Similarly, if the joint angles of the components are known, the extension and retraction of each hydraulic cylinder can be calculated through geometric relationships, and the conversion relationship between the two can be solved to establish the correlation between the control drive variable and the rotational variable of the structural component. The connection between the hydraulic cylinder and the structural component is relatively simple, and analytical geometry methods can be used to analyze and solve the relationship between the extension and retraction length of the hydraulic cylinder and the rotational angle between the joints.

[0093] In one embodiment of the present invention, step 200, which involves acquiring terrain information in real time, includes:

[0094] 210. Acquire point cloud data generated by lidar scanning of the ski slope ahead.

[0095] Specifically, the terrain perception module continuously receives raw point cloud data streams from the multi-line lidar.

[0096] 220. Organize the point cloud data and segment ground points and non-ground points according to dynamically adjusted discrimination thresholds.

[0097] In this step, the terrain perception module continuously receives raw point cloud data streams from the multi-line lidar. First, data preprocessing is performed, filtering out the vehicle's own point cloud using the radar's installation location information. Next, the point cloud is ordered, arranging it into a regular data structure based on the lidar beam number and horizontal rotation angle corresponding to each point, arranged according to the scan lines. Then, adaptive ground segmentation is performed: instead of using fixed thresholds, the algorithm dynamically calculates horizontal distance and elevation difference thresholds between the current scan line and its adjacent lines based on the radar's known installation height, the elevation angle of the current scan line, and a real-time updated slope estimate. By comparing these dynamic thresholds with the actual geometric relationships between adjacent point clouds, points belonging to continuous snow track surfaces can be accurately separated from points belonging to snowdrifts or other protrusions.

[0098] 230. Based on the segmented ground point cloud data, calculate the slope and distance of the ski slope ahead.

[0099] In this step, based on the segmented clean ground point cloud, multiple continuous scan lines in the area directly in front of the vehicle are selected. The slope is obtained by calculating the ratio of the height difference of the points on these lines to the horizontal distance, and the slope distance is calculated using the coordinates of the foremost ground point.

[0100] In one embodiment of the present invention, the dynamically adjusted discrimination threshold in step 220 is determined based on the geometric installation parameters of the lidar, the angle information of the scanning beam, and the real-time estimated terrain slope information.

[0101] Specifically, this dynamic thresholding mechanism is used to improve ground segmentation accuracy. The fixed installation height of the lidar and the fixed pitch angle of each scan line are known geometric parameters. The algorithm introduces a real-time estimated slope angle as a key variable. When determining whether two points on adjacent scan lines belong to the same ground, the expected elevation difference between the ground points is no longer zero, but is related to the horizontal distance between the two points and the tangent of the current slope angle. Therefore, the discrimination threshold is adaptively adjusted based on the real-time estimated slope angle. When the slope of the ski run is large, the threshold is widened accordingly to prevent steep slopes from being misclassified as obstacles; when the slope is small, the threshold is tightened to avoid misclassifying low snowdrifts as ground. This dynamic adjustment mechanism ensures accurate ground point extraction on ski runs with different slopes, providing a reliable data foundation for subsequent slope calculations.

[0102] In one embodiment of the present invention, calculating the slope of the ski slope ahead in step 230 includes:

[0103] 231. Select multiple adjacent scan line data from the ground point cloud.

[0104] To obtain robust and representative slope values, a multi-line fusion calculation strategy is employed. From the segmented ground point cloud, a specified number of continuous scan lines directly in front of the snow groomer are selected, typically including line bundles in the left, center, and right directions.

[0105] 232. Calculate the slope information representing topographic changes in each scan line data.

[0106] For each selected scan line, a series of ordered ground points are extracted, the ratio of the height difference between two adjacent points to the horizontal distance is calculated, multiple local slopes are obtained, and the average value of these local slopes is taken as the slope estimate of the scan line direction.

[0107] 233. The slope information calculated from multiple scan lines is fused and processed to output the slope value.

[0108] Specifically, the slope estimates calculated from three or more scan lines (left, center, and right) are averaged or weighted to output a final ski slope value. This method effectively smooths out random errors caused by sparse point clouds or minor local undulations, improving the overall accuracy and robustness of slope sensing.

[0109] ① Calculate the horizontal angle of the original point cloud. The y-axis is directly in front of the vehicle, and its horizontal angle is defined as 0. The angle increases sequentially as the vehicle rotates clockwise. Calculate the horizontal angle of the original point cloud using the formula.

[0110] Formula 9

[0111] In the formula, For point Projected coordinates in the XOY plane For point The horizontal angle, k is an integer.

[0112] ② Point clouds at the same horizontal angle but on different vertical scan lines differ in their distance from the radar center. Arranging them from smallest to largest distance yields an ordered arrangement of the lidar point clouds vertically from the first scan line to the last.

[0113] An adaptive ground segmentation algorithm is adopted, which can dynamically adjust the discrimination threshold according to the real-time estimated slope of the snow track, thereby accurately classifying point clouds into ground points and non-ground points such as snowdrifts.

[0114] Finally, based on the segmented ground point cloud, by analyzing the spatial distribution of the point cloud along a specific scan line, and using a method of successively calculating and fusing multi-line information, the slope angle of the snow track within a predetermined distance in front of the snow groomer and the distance between the starting point of the slope and the vehicle are calculated in real time.

[0115] Adaptive horizontal distance threshold Local height threshold and global height threshold See the formula.

[0116] Formula 10

[0117] Formula 11

[0118] Formula 12

[0119] As can be seen from the formula, we only need to set the installation height H of the lidar and a fixed slope value in advance, and we can adaptively adjust the local height and global height threshold of the two scanning circles before and after according to the vertical angle of the adjacent beams, thereby improving the real-time performance of the algorithm and the accuracy of ground segmentation.

[0120] Currently, this algorithm is only suitable for surfaces with relatively small slopes, while ski resorts have significant slope variations, reaching up to 25 degrees. If a fixed slope value is set in advance, and this value is too small relative to the actual slope, the ski slope surface will be mistaken for an obstacle and filtered out. Conversely, if the slope value is set too large relative to the actual slope, snowdrifts on the slope will be mistaken for the ground and retained. Both situations result in poor ground segmentation, leading to inaccurate slope extraction.

[0121] By observing the ski resort environment on-site, the ski slopes were basically flat, with snow mounds, but virtually no other obstacles. Therefore, the slope of the ski slope could be calculated once before segmenting the ground as the initial slope, so that the slope set in the algorithm is close to the actual slope, thereby avoiding undersegmentation and oversegmentation.

[0122] The following describes the specific process of point cloud segmentation. When judging the first scan line of each scan plane, if it is less than the global threshold, it is a ground point; otherwise, it is a non-ground point.

[0123] When the current point is a ground point, the system determines whether the next ray point is a ground point based on an adaptive horizontal distance threshold and a local height threshold. If the current point is a ground point and the horizontal distance to the next point is less than the horizontal distance threshold, then the next point is an obstacle point. If the current point is not a ground point, the system determines whether the next ray point is a ground point based on an adaptive global height threshold. If the current point is not a ground point, but the next point simultaneously satisfies the condition that the height difference is less than the local height threshold and the height is less than the global height threshold, then the next point is a ground point; otherwise, it is a non-ground point.

[0124] After the ground segmentation is completed, the slope can be calculated and the distance output. To reduce the influence of random errors during slope calculation, a successive difference method is used. Points on N scan lines in front of the snow groomer are selected, and the average slope between N points on the left, center, and right lines is calculated separately. Then, the average slope of the three lines is calculated to improve calculation accuracy, as shown in the formula. When outputting the distance, the y-coordinate value of the first point on the center line is selected as the distance of the slope in front of the snow groomer. Figure 3 This is a schematic diagram for slope calculation. The lines of different colors represent point clouds at different heights scanned. Figure 4 For processing flowcharts.

[0125] Formula 13

[0126] Formula 14

[0127] Formula 15

[0128] In one embodiment of the present invention, step 300, which calculates the target control quantity of the drive system based on the kinematic model, includes:

[0129] 310. Based on the target pose of the shovel tip, calculate the required rotation angle of each joint through inverse kinematics.

[0130] This step realizes the transformation from task space target to pose space command and then to drive space command. First, the task requirement of "shovel tip parallel to the ski slope" is quantified into the target pose angle of the shovel tip by combining it with the real-time sensed ski slope. At the same time, the target height or position coordinates of the shovel tip are set according to the operational requirements. These target pose parameters are input into the inverse kinematics model. Based on the mathematical relationship derived from the forward kinematics, the inverse kinematics model solves in reverse the direction to determine the angle values ​​that the front connecting frame and the swing frame joints need to rotate to achieve the target pose.

[0131] 320. Based on the geometric relationship between the joint rotation angle and the extension / retraction of the drive mechanism, the target extension / retraction of each drive mechanism is obtained.

[0132] Specifically, a second conversion is performed: since the snowplow is ultimately driven by a hydraulic cylinder, the joint angle needs to be converted into the hydraulic cylinder length. Based on the fixed geometric triangle formed by the front connecting frame, swing frame, chassis, and hydraulic cylinder hinge points, trigonometric relationships can be used to uniquely calculate the target extension length of the piston rods of the corresponding lifting cylinder and tilting cylinder from the known joint angles. This target length is the control command directly sent to the hydraulic drive system.

[0133] Let the coordinates of point F, the tip of the snow groomer's shovel, be... The angle between the shovel tip and the ground is The extension and retraction length of the hydraulic cylinder is then calculated.

[0134] Based on the known coordinates of the snowplow tip described above, we can obtain:

[0135] Formula 16

[0136] Depend on Figure 2 It can be seen that: , ,and:

[0137] Formula 17

[0138] Formula 18

[0139] Organized , The expression is:

[0140] Formula 19

[0141] Formula 20

[0142] In one embodiment of the present invention, the adaptive control strategy in step 400 includes:

[0143] 410. A reference model is preset to characterize the desired dynamic performance.

[0144] The reference model adaptive control strategy is used to address the time-varying, nonlinear, and disturbance-dependent issues of parameters in snowplow hydraulic systems. First, a linear, time-invariant reference model is designed, whose transfer function possesses ideal dynamic characteristics such as fast response, no overshoot, and no steady-state error, representing our performance expectations for the hydraulic cylinder position control system. The actual controlled object is the valve-controlled hydraulic cylinder system of the snowplow.

[0145] 420. Use parameter adaptive regulation law for control.

[0146] Specifically, an adaptive controller can be constructed, which is essentially an adjustable system. During control, both the reference model and the adjustable system receive the same target command input.

[0147] 430. By comparing the error between the output of the controlled system and the output of the reference model, the control parameters are dynamically adjusted so that the controlled system tracks the dynamic characteristics of the reference model.

[0148] The adaptive mechanism continuously monitors the error between the actual output of the adjustable system and the ideal output of the reference model. Once the error becomes non-zero, it indicates that the current controller parameters have failed to achieve the ideal performance of the controlled object. At this point, the adaptive mechanism automatically adjusts the internal parameters of the adjustable system based on a parameter update law rigorously derived from Lyapunov stability theory, thereby changing its output. The ultimate goal is to force the tracking error to converge to zero. This allows the actual hydraulic system to approximate the dynamic performance of the ideal reference model, exhibiting good response speed and anti-interference capabilities.

[0149] Suppose the differential equation of the controlled object is as follows:

[0150] Formula 21

[0151] The differential equation of the reference model is as follows:

[0152] Formula 22

[0153] In the above equation, p is the differential operator. The expression is given below:

[0154] Formula 23

[0155] Formula 24

[0156] Formula 25

[0157] In the controlled object, a0, a1…a n-1 ,b0,b1,…b m-1 ,b m All parameters are unknown, A m (p) is a stable polynomial. We make the following assumption:

[0158] a) Amplification factor of the controlled object: b > 0;

[0159] b) B(p) is a stable polynomial;

[0160] c) m ≥ q;

[0161] d) The output differential value of the controlled object { It can be used.

[0162] When the standard input is u m Given y(t), the task of the adaptive controller is to dynamically determine the control input u(t) so that the output y(t) of the controlled object gradually approaches the output y(t) of the reference model. m (t).

[0163] To derive the parameter adjustment law and the adaptive control law, we first assume... Satisfy the following formula:

[0164] Formula 26

[0165] in, It is a polynomial of order less than n-1, because The parameters are unknown, therefore The parameters are also unknown. The expression for the adaptive error is: Substitute this into the following formula:

[0166]

[0167] The formula is rearranged to obtain the adaptive error expression:

[0168] Formula 27

[0169] Introducing a stable polynomial H(p) of order m into the above formula:

[0170] Formula 28

[0171] Choose a polynomial of order nm-1 ,make It is a strictly positive real function, and the following assumptions are made:

[0172] Formula 29

[0173] Combining equations 28 and 29, we get:

[0174] Formula 30

[0175] In the formula, It is the fundamental error equation for reference model adaptive control. It is a differential term with output error. A state-variable filter is introduced here, and its output signal... for:

[0176] Formula 31

[0177] In the formula The system has m+n+1 state filters. Assume:

[0178] Formula 32

[0179] In the formula, S1(p) is an m-1 order polynomial with a total of m unknowns, r i (i=1,2,…,ρ) are ρ unknowns, e d The expression for (t) is represented in the following form:

[0180] Formula 33

[0181] In the formula, It is an unknown parameter vector of order n+m+1. This corresponds to the unknown parameters of B(p)-H(p); Corresponding to Unknown parameters; The corresponding parameter is the unknown parameter. (i=0,1,…ρ;ρ=nm-1); The corresponding parameter is the unknown parameter. .

[0182] The expression for the control input u(t) of the controlled system is:

[0183] Formula 34

[0184] e d (t) can be rearranged into the following form:

[0185] Formula 35

[0186] In the formula, K(t) is a vector of adjustable parameters of order n+m+1, and the error vector of order n+m+1 is... The expression is as follows:

[0187] Formula 36

[0188] e d The expression for (t) can be rearranged into the following form:

[0189] Formula 37

[0190] because It is strictly true; according to the equilibrium point stability theorem, we take... The expression is as follows:

[0191] Formula 38

[0192] The expression for the limit of the adaptive error e(t) approaching 0 is as follows:

[0193] Formula 39

[0194] Parameter adjustment rules The expression is as follows:

[0195] Formula 40

[0196] according to Where D(p) is a stable polynomial, we have:

[0197] Formula 41

[0198] This forms the adaptive control system of the reference model.

[0199] According to the closed-loop transfer function of the snowplow hydraulic control system, the order of the numerator polynomial is m=0, and the order of the denominator polynomial is n=3. To ensure the adaptive system remains stable in the Lyapunov sense, an m-th order polynomial H(p)=1 and an nm-1-th order polynomial D(p)=p are used. 2 +140p+10000, and get:

[0200] Formula 42

[0201] Equation 42 is strictly positive. To simplify the control algorithm, the generalized output error is taken as follows:

[0202] Formula 43

[0203] Based on the knowledge of adaptive theory, we know that m=0, H(p)=1, and the output signal of the state filter is x. i (t), (i=1,2,…n+m+1), in the hydraulic system of this paper, the number of filters is n+m+1=4, and the output signal is as follows:

[0204] Formula 44

[0205] The adaptive parameter regulator is as follows:

[0206] Formula 45

[0207] The input of the controlled object is:

[0208] Formula 46

[0209] Because the response speed and stability of the snowplow hydraulic system are still somewhat lacking, a reference model was introduced to improve the characteristics of the hydraulic system and give it good dynamic performance. The selected reference model is G. m The expression for (s) is shown below:

[0210] Formula 47

[0211] To obtain the adaptive generalized error, an adaptive control system simulation model can be easily constructed using Simulink. The transfer function needs to undergo an inverse Laplace transform. The differential equations of the hydraulic system model of the snow groomer and the reference model are as follows:

[0212] Formula 48

[0213] After three integration operations, the hydraulic control system of the controlled system, i.e., the snowplow, and the acceleration, velocity, and displacement of the parameter model can be obtained respectively.

[0214] In one embodiment of the present invention, the parameter adaptive adjustment law in step 420 includes:

[0215] 421. Construct a filter auxiliary signal vector containing the system input and output signals.

[0216] To obtain sufficient information to adjust all unknown or time-varying parameters in the controller, a set of state variable filters needs to be constructed. These filters process the system's control input signal and the controlled object's output signal, generating a series of filtered auxiliary signals. These auxiliary signals together form an auxiliary signal vector.

[0217] 422. Design a parameter update law based on stability theory, and adjust the control parameters in real time according to the tracking error and auxiliary signal vector.

[0218] Specifically, the parameter update law is designed as an integral form of adaptive law, whose inputs are the system's generalized output error and this auxiliary signal vector. The mathematical form of this update law guarantees the global asymptotic stability of the entire adaptive closed-loop system. In actual operation, the algorithm runs continuously, calculating and updating the controller's gain and other parameters in real time based on the current tracking error and auxiliary signal vector. Therefore, without manual intervention, the control system maintains excellent tracking performance even under conditions such as changes in the hydraulic oil's elastic modulus and load fluctuations. The adaptive model simulation of the snowplow: Based on the differential equations, the controlled system and reference system are built in Simulink. See the simulation model diagram for details. Figures 6 to 7 Please refer to the simulation results diagram. Figures 8 to 13 .

[0219] pass Figures 8 to 10 It can be seen that after adding various signals, the output signal value of this adaptive control system follows the reference model very well, and the error e in the system is very small. m The signal approaches zero. The curve also shows that the signal response is fast, reaching a steady state in about 0.1 seconds. The dynamic response process is oscillating, and compared to the input signal, the signal does not exceed the amplitude, achieving the design target.

[0220] The control effect analysis was conducted using step functions and sine functions as inputs, and the results show that the system has good dynamic response and tracking performance. Based on the length changes of the lifting cylinder and the tilting cylinder obtained from the kinematic simulations in the previous sections, the cylinder extension / retraction amount starting from the middle position can be calculated. This extension / retraction amount is used as the input to the control system to analyze the tracking effect of the adaptive control system.

[0221] The simulation results are obtained by using the changes in the lifting cylinder and the tilting cylinder as inputs to the control system, respectively. Figure 11 As shown, their output error curves are as follows: Figure 12As shown in the figure, the hydraulic system responds quickly with almost no delay when the input is the change in the lifting cylinder. Before 1.1 seconds, the tracking error decreases from within 1.22 mm to within 0.2 mm. Between 1.1 and 1.8 seconds, there is a tracking error of -1 mm to +1.5 mm. Between 1.8 and 4.2 seconds, the tracking error is between -0.024 mm and 0 mm. After 4.2 seconds, there is a tracking error of -1.95 mm to +0.8 mm. When the input is the change in the tilting cylinder, the error is between -0.8 mm and +0.6 mm in the first 0.7 seconds, and within 0 to 0.004 mm after 0.7 seconds. Therefore, the maximum tracking error of the snowplow lifting cylinder is 1.95 mm, and the maximum tracking error of the tilting cylinder is 0.8 mm. Compared to a hydraulic system without adaptive control, this represents a significant improvement in both control accuracy and system response.

[0222] Figure 13 The time-domain characteristic curve and error curve of the hydraulic system, with the input being the change in the tilting cylinder length, show an error of -0.8 mm to +0.6 mm in the first 0.7 seconds, and a tracking error within 0 to 0.004 mm after 0.7 seconds. From the above, it can be seen that the maximum tracking error of the snowplow lifting cylinder is 1.95 mm, and the maximum tracking error of the tilting cylinder is 0.8 mm. Compared with the hydraulic system without adaptive control, this represents a significant improvement in both control accuracy and system response.

[0223] In one embodiment of the present invention, the method enables the snowplow to automatically adjust the tip posture under various working conditions where the relative posture of the vehicle body and the snow track changes, so as to maintain a predetermined relative relationship between the tip and the snow track surface.

[0224] Specifically, the complete control system constructed using this method possesses autonomous decision-making capabilities to handle typical working conditions. Specifically, during the snow groomer's operation, the system can automatically identify and handle the following three typical scenarios: when both the vehicle body and the snowplow are on a flat section of road, it maintains the snowplow tip's baseline posture; when the vehicle body has not yet entered the slope but the snowplow tip has already contacted it, the system can immediately adjust the snowplow tip's downward angle in advance based on the perceived slope ahead, preventing the snowplow tip from inserting too deeply into the snow; when the entire vehicle body begins to enter the slope, the system can dynamically and smoothly adjust the snowplow tip's posture from a downward state back to a parallel state adapted to the new vehicle body posture during the uphill process. Through this proactive sensing and continuous adjustment, regardless of how the relative posture of the vehicle and the snow slope changes, the system can automatically and continuously adjust the actions of each hydraulic cylinder to ensure that the snowplow tip and the snow slope surface always maintain a predetermined parallel contact, thereby achieving stable and uniformly thick snow pushing operations.

[0225] Figure 14This is a block diagram illustrating an adaptive coordination control device for snow groomers and snowplows according to an exemplary embodiment. The device is used in an adaptive coordination control method for snow groomers and snowplows. (Refer to...) Figure 14 The device includes: a model building module 1501, a terrain acquisition module 1502, an attitude determination module 1503, and a control module 1504. In one embodiment of the present invention, a snow groomer snowplow adaptive coordination control device is provided, which includes:

[0226] Model building module 1501 is used to build the kinematic model of the snow shovel and determine the mapping relationship between the shovel tip posture and the motion parameters of each drive joint;

[0227] This module can be integrated into the vehicle controller or the host industrial computer. Its function is realized by storing and executing kinematic modeling and calculation algorithms. It stores all the key geometric parameters of the snowplow mechanism and can derive and calculate the mathematical relationship between the snowplow tip posture and the front connecting frame rotation angle and the swing frame rotation angle, either offline or online, based on multibody kinematics theory and homogeneous transformation method, providing an accurate mathematical model for the entire control system.

[0228] The terrain acquisition module 1502 is used to acquire real-time terrain information of the snow track in front of the snow groomer, including slope and distance information.

[0229] This module includes a lidar sensor and its driving circuit, and a point cloud processing unit. The lidar is fixedly mounted on the roof of the vehicle and is responsible for collecting raw point clouds. The point cloud processing unit runs algorithms for point cloud ordering, adaptive ground segmentation, and slope calculation to process the raw data in real time, and finally outputs the slope angle and distance information of the ski slope ahead, which is then sent to the main controller.

[0230] The attitude determination module 1503 is used to determine the target attitude of the snow shovel based on the slope in the terrain information, and to calculate the target control quantity of the drive system based on the kinematic model.

[0231] This module typically serves as the core decision-making software component of the main controller. It receives slope information from the terrain acquisition module and, combined with preset operational rules such as keeping the shovel tip parallel, determines the target pose of the shovel tip. Subsequently, it calls the inverse kinematics function provided by the model building module to calculate the required joint target rotation angle. Then, through built-in geometric transformation relationships, it finally calculates the target extension and retraction length commands for the lifting cylinder and the tilting cylinder.

[0232] The control module 1504 is used to coordinate and control the drive system of the snowplow based on the target control quantity according to the adaptive control strategy, so that the actual posture of the snowplow tracks the target posture to adapt to changes in the snow track.

[0233] This module includes an adaptive control algorithm unit and an electro-hydraulic proportional valve drive unit. The algorithm unit implements the adaptive control law of the reference model, receives the target control quantity command, and simultaneously obtains feedback on the actual extension and retraction of each hydraulic cylinder through displacement sensors. After calculation by the adaptive algorithm, a real-time control voltage signal is generated. The drive unit amplifies this voltage signal and drives the electro-hydraulic proportional valves controlling the lifting cylinder and the tilting cylinder, thereby precisely controlling the hydraulic flow and direction. This achieves rapid, accurate, and coordinated control of the hydraulic cylinder extension and retraction movements, ultimately ensuring that the snowplow's actual posture accurately tracks the desired target.

[0234] The device provided in this embodiment integrates high-precision kinematic modeling, real-time terrain perception based on lidar, and robust adaptive control technology, enabling intelligent and precise snow grooming operations. It effectively solves problems such as uneven snow removal depth, snow jamming, or missed areas caused by long visibility and judgment delays in traditional manual operations. It can automatically and quickly adjust the snow plow's posture to follow changes in the slope of the snow track, significantly improving the quality and efficiency of snow track leveling operations, reducing operator workload, and enhancing the adaptability and operational stability of the snow groomer under complex and variable snow conditions.

[0235] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0236] In this invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0237] It should be understood that, in various embodiments of the present invention, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0238] 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.

Claims

1. An adaptive coordinated control method for snow groomer snowplows, characterized in that, Includes the following steps: Establish a kinematic model of the snow shovel and determine the mapping relationship between the shovel tip posture and the motion parameters of each drive joint; Real-time acquisition of terrain information of the snow track in front of the snow groomer, including slope and distance information; Based on the slope in the terrain information, the target posture of the snow shovel is determined, and the target control quantity of the drive system is calculated based on the kinematic model. Based on an adaptive control strategy, the drive system of the snowplow is coordinated and controlled according to the target control quantity, so that the actual posture of the snowplow tracks the target posture to adapt to changes in the snow track.

2. The method according to claim 1, characterized in that, Establishing the kinematic model includes: Based on the theory of multibody kinematics, a coordinate system is established at the key hinge points of the kinematic chain of the snow shovel mechanism. By using homogeneous coordinate transformation, the functional expression of the shovel tip pose with respect to the rotation angles of each joint is derived.

3. The method according to claim 1, characterized in that, Real-time acquisition of the terrain information includes: Acquire point cloud data generated by lidar scanning the ski slope ahead; The point cloud data is organized, and ground points and non-ground points are segmented according to dynamically adjusted discrimination thresholds; Based on the segmented ground point cloud data, the slope and distance of the ski slope ahead are calculated.

4. The method according to claim 3, characterized in that, The dynamically adjusted discrimination threshold is determined based on the geometric installation parameters of the lidar, the angle information of the scanning beam, and the real-time estimated terrain slope information.

5. The method according to claim 3, characterized in that, Calculating the slope of the ski run ahead includes: Select multiple adjacent scan line data from the ground point cloud; Calculate the slope information representing topographic changes in each scan line data; The slope information calculated from multiple scan lines is fused and processed to output the slope value.

6. The method according to claim 1, characterized in that, The target control quantity of the drive system is calculated based on the kinematic model, including: Based on the target pose of the shovel tip, the required rotation angle of each joint is calculated through inverse kinematics. Based on the geometric relationship between the joint rotation angle and the extension / retraction of the drive mechanism, the target extension / retraction of each drive mechanism is obtained.

7. The method according to claim 1, characterized in that, The adaptive control strategy includes: A reference model is pre-defined to characterize the desired dynamic performance; Control is achieved using a parameter adaptive regulation law; By comparing the error between the output of the controlled system and the output of the reference model, the control parameters are dynamically adjusted so that the controlled system tracks the dynamic characteristics of the reference model.

8. The method according to claim 7, characterized in that, The adaptive adjustment law for the parameters includes: Construct a filter auxiliary signal vector containing the system input and output signals; The parameter update law is designed based on stability theory, and the control parameters are adjusted in real time according to the tracking error and the auxiliary signal vector.

9. The method according to claim 1, characterized in that, The method enables the snowplow to automatically adjust the tip posture under various working conditions where the relative posture of the snowplow body and the snow track changes, so as to maintain the predetermined relative relationship between the tip and the snow track surface.

10. An adaptive coordination control device for snow groomer snowplows, characterized in that, The device includes: The model building module is used to build the kinematic model of the snow shovel and determine the mapping relationship between the shovel tip posture and the motion parameters of each drive joint; The terrain acquisition module is used to acquire real-time terrain information of the snow track in front of the snow groomer, including slope and distance information; The attitude determination module is used to determine the target attitude of the snow shovel based on the slope in the terrain information, and to calculate the target control quantity of the drive system based on the kinematic model. The control module is used to coordinate and control the drive system of the snowplow based on the target control quantity according to the adaptive control strategy, so that the actual posture of the snowplow tracks the target posture to adapt to changes in the snow track.

Images