A navigation and control method for a four-wheel drive differential-driven heavy-duty AGV

Abstract

This invention discloses a navigation and control method for a four-wheel drive differential-driven heavy-duty AGV, comprising the following steps: treating the two front drive groups of the AGV as a first drive system and the two rear drive groups as a second drive system; calculating the center position coordinates of the first and second drive systems; establishing a local coordinate system; calculating the lateral offset of the two positions from the global reference path; obtaining the center deviation of the first and second drive systems based on the lateral offset; calculating control signals for the steering angular velocities of the first and second drive systems; calculating a speed control signal; and adjusting the front and rear position attitudes of the AGV based on the control signals and speed control signals for the steering angular velocities of the first and second drive systems. This application achieves independent and coordinated control of the front and rear ends of the heavy-duty AGV, improving trajectory tracking accuracy and attitude control stability.

Description

Technical Field

[0001] This invention belongs to the field of AGV navigation and control technology, specifically relating to a navigation and control method for a four-wheel drive differential-driven heavy-duty AGV. Background Technology

[0002] Automated Guided Vehicles (AGVs), as core equipment for material handling and equipment docking in the field of industrial automation, have their applicable scenarios and operational precision directly determined by their drive method and navigation control technology. Currently, the mainstream drive forms of AGVs are mainly divided into two categories: differential drive and Mecanum wheel drive. Mecanum wheel drive, with its advantage of flexible multi-directional movement, is widely used in light-load, frequently turning scenarios. However, due to the limitations of its wheel structure, its load-bearing capacity is relatively weak and cannot meet the needs of heavy-load scenarios. On the other hand, differential drive AGVs, due to their stable drive structure and outstanding load-bearing performance, have become the core choice in heavy-load applications.

[0003] In the navigation and motion control technology of multi-group differential drive AGVs, existing solutions typically employ two kinematic models: the traditional Ackerman control kinematic model and the small-radius turning kinematic model for four-wheel drive synchronous turning. However, heavy-duty AGVs, due to their functional requirements, generally suffer from inherent problems such as large size and poor turning flexibility. Simultaneously, the navigation accuracy of AGVs is highly dependent on positioning accuracy. In factory environments, the mainstream positioning method is SLAM positioning, whose positioning accuracy typically only reaches ±30mm, resulting in unavoidable positioning data fluctuations. The combination of these factors leads to heavy-duty AGVs using the traditional Ackerman control model or the small-radius turning kinematic model being prone to large attitude deviations and tail-wagging during position adjustments. This problem is particularly prominent in narrow working areas (such as workshop aisles and equipment gaps) or scenarios requiring high-precision control (such as docking with automated production lines and loading / unloading precision components). This not only makes precise docking with other automated equipment difficult but may also affect the continuity of the production process and restrict the overall efficiency of the automated production line.

[0004] To address the high-precision control requirements of heavy-duty AGVs, some traditional solutions propose installing magnetic sensors at both ends of the AGV and deploying magnetic strips along its preset travel path. The magnetic sensors read the signals from the magnetic strips to determine the AGV's offset relative to the travel path, and then calculate and output control signals to achieve high-precision navigation control. However, this solution has significant drawbacks in practical applications, as follows:

[0005] (1) Additional purchase of magnetic sensors and supporting signal processing components is required, which increases the hardware cost of AGV equipment and is not conducive to large-scale promotion and application.

[0006] (2) The magnetic strips deployed on the ground are easily crushed and damaged by other operating vehicles (such as forklifts and transfer vehicles) in the factory area. They need to be inspected, repaired or replaced regularly, which not only increases the workload of later maintenance, but also affects the production progress due to maintenance shutdowns.

[0007] (3) For exhibition hall-style workshops, high-end manufacturing plants and other scenarios with high requirements for the aesthetics of the floor, laying magnetic strips on the floor will destroy the overall visual effect and conflict with the aesthetic requirements of the scenario, and cannot meet the usage needs of diverse scenarios.

[0008] (4) In a strong magnetic field environment (such as a workshop equipped with large electromagnetic equipment or a metal processing scene), the strong external magnetic field will interfere with the signal generated by the magnetic strip, causing the magnetic sensor reading to be distorted and the function to fail, making it impossible to achieve stable positioning and control, and losing the high-precision control effect.

[0009] In summary, existing navigation and control technologies for heavy-duty AGVs still face significant technical bottlenecks in meeting core requirements such as high precision, low cost, and high adaptability. A new navigation and control solution is urgently needed to address the shortcomings of existing technologies and meet the high-precision and high-reliability control requirements of AGVs in heavy-duty scenarios. Summary of the Invention

[0010] In view of the technical problems mentioned in the background, the purpose of this invention is to provide a navigation and control method for a four-wheel drive differential-driven heavy-duty AGV.

[0011] To achieve the objectives of this invention, the technical solution provided by this invention is as follows:

[0012] A navigation and control method for a four-wheel drive differential-driven heavy-duty AGV includes the following steps:

[0013] Step S1: Consider the two sets of drives on the front side of the AGV as the first drive unit and the two sets of drives on the rear side as the second drive unit. Calculate the coordinates of the center position P1 of the first drive unit. x 1， y 1) Coordinates of the overall center position P2 of the second drive ( x 2， y 2);

[0014] Step S2: Establish a local coordinate system F with the global reference path and the lateral deviation direction as the coordinate axes;

[0015] Step S3: Calculate the lateral offsets of positions P1 and P2 from the global reference path. d 1 and d 2 ;

[0016] Step S4: Based on the lateral offset d 1 、d 2 The magnetic navigation deviation data, namely the first drive overall center deviation Δ1 and the second drive overall center deviation Δ2, are obtained respectively.

[0017] Step S5: Calculate the steering angular velocity control signal of the first drive assembly and the steering angular velocity control signal of the second drive assembly based on the center deviation Δ1 of the first drive assembly and the center deviation Δ2 of the second drive assembly.

[0018] Step S6: Calculate the speed control signal;

[0019] Step S7: Adjust the front and rear positions and attitudes of the AGV according to the steering angular velocity control signal of the first drive system and the steering angular velocity control signal and speed control signal of the second drive system.

[0020] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0021] (1) In this application, the front and rear drives of the AGV are regarded as two separate entities. The front and rear drives are controlled independently. Their path offsets are calculated by laser SLAM positioning, and their steering and speed are controlled independently. This achieves flexible adjustment and perfectly solves the problem of inflexible turning and easy tail swing of the traditional Ackerman model. It realizes independent and coordinated control of the front and rear ends of the heavy-duty AGV, which greatly improves the accuracy of trajectory tracking and attitude adjustment.

[0022] (2) The independent steering control of the front and rear of this application enables the AGV to make turns with smaller radii and more complex posture adjustments, which greatly enhances its adaptability in complex workshop environments and significantly improves the turning and docking capabilities of large heavy-duty AGVs in narrow spaces.

[0023] (3) This application uses a magnetic navigation deviation calculation module to "virtually" generate magnetic navigation deviation data through laser SLAM positioning and algorithms. This not only reduces hardware procurement costs, but also eliminates the high costs and downtime of laying, maintaining and replacing magnetic strips, while avoiding interference problems in strong magnetic field environments and eliminating the dependence on physical magnetic strips and magnetic sensors;

[0024] (4) This application uses a magnetic navigation deviation calculation module to replace the deviation signal of the physical magnetic sensor with geometric information. It does not require physical ground markings, thus keeping the ground clean and beautiful. It does not rely on magnetic signals, thus naturally being immune to electromagnetic interference, which broadens the application scenarios. It is suitable for showroom-style workshops where the ground is aesthetically pleasing, as well as industrial environments with strong magnetic field interference.

[0025] (5) Compared with the traditional model, the new control strategy of this application can distribute the driving force and steering angle of each wheel more smoothly and accurately, reducing mechanical stress and tire wear. It extends the service life of the equipment; through precise algorithm control, it avoids severe tire slippage and wear during cornering. Attached Figure Description

[0026] Figure 1 A schematic diagram illustrating the adjustment of AGV posture under traditional control mode;

[0027] Figure 2 A schematic diagram illustrating the adjustment of AGV pose using the scheme described in this application;

[0028] Figure 3 This is a schematic diagram showing the location of the kinematic center of the AGV.

[0029] Figure 4 This is a schematic diagram showing the positions of P1 and P2 in this application.

[0030] Figure 5 This is a schematic diagram of the AGV control trajectory under the traditional control mode;

[0031] Figure 6 A schematic diagram of the AGV control trajectory using the scheme in this application. Detailed Implementation

[0032] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0033] It should be noted that the AGV described in this invention is a typical four-wheel drive differential structure, with four sets of drive wheels arranged symmetrically on the left and right sides of the front and rear of the vehicle, with the longitudinal centerline of the vehicle as the axis of symmetry. This invention belongs to AGV navigation and control technology and is suitable for working conditions with high requirements for the accuracy of the travel trajectory, such as scenarios involving the transfer of stirrups and docking of robotic arms. The solution in this application does not rely on magnetic sensors. Instead, it uses laser SLAM positioning and AGV size parameters to obtain a magnetic-like navigation deviation through specific mathematical calculations, and introduces a magnetic-like navigation deviation calculation module to achieve high-precision control of the AGV. The "magnetic-like navigation deviation calculation module" introduced in this invention is not a physical magnetic field sensor, nor does it perform magnetic or electromagnetic field simulation. This module is a software / algorithm module, and its function is to generate a lateral deviation (deviation signal) consistent with that of a traditional physical magnetic sensor in a magnetic strip navigation scenario in a control sense, as an input to the controller.

[0034] This module calculates the geometric lateral deviation between the AGV's kinematic center or the center point of the drive system and a preset trajectory centerline (similar to a virtual magnetic strip), and performs proportional transformation and amplitude limiting according to the controller calibration coefficients, thereby outputting a deviation signal for path tracking control. Since this output is semantically equivalent to the output of a physical magnetic sensor in terms of controller interface, it is called a "magnetic-like navigation deviation." Figure 1 A schematic diagram illustrating the adjustment of AGV posture under traditional control mode; Figure 2 A schematic diagram illustrating the adjustment of AGV pose using the scheme described in this application; Figure 3 This is a schematic diagram showing the location of the kinematic center of the AGV. Figure 4 This is a schematic diagram showing the positions of P1 and P2 in this application. Figure 5 This is a schematic diagram of the AGV control trajectory under the traditional control mode; Figure 6 A schematic diagram of the AGV control trajectory using the scheme in this application.

[0035] This embodiment provides a navigation and control method for a four-wheel drive differential-driven heavy-duty AGV, including the following steps:

[0036] Step S1: Consider the two front drive groups of the AGV as a whole, and the two rear drive groups as a whole. Treat the two front drive groups as the first drive group and the two rear drive groups as the second drive group, and calculate the coordinates of the center position P1 of the first drive group. x 1， y 1) Coordinates of the overall center position P2 of the second drive ( x 2， y 2);

[0037] The "first drive unit" of this application consists of two sets of symmetrically arranged drive wheel units on the front side of the AGV; the "second drive unit" consists of two sets of symmetrically arranged drive wheel units on the rear side of the AGV. Since the drive wheels are symmetrical front-to-back and left-to-right, their physical center position can be determined by the midpoint of the line connecting the center points of the two drive wheels.

[0038] It should be noted that, as Figure 4 As shown, this application considers the two sets of drives on the front side of the AGV as a whole, and the two sets of drives on the rear side as a whole.

[0039] Front point coordinates P1 = vehicle center P + (wheel_base / 2) × [cos(θ), sin(θ)];

[0040] Rear point coordinates P2 = vehicle center P - (wheel_base / 2) × [cos(θ), sin(θ)];

[0041] Where wheel_base represents the wheelbase; θ represents the AGV orientation. This formula is not an arbitrarily set ideal point, but rather an engineering equivalent expression in the longitudinal axis direction of the vehicle based on the aforementioned physical center point: In a four-wheel drive symmetrical arrangement, the midpoint of the drive wheel center point must fall on the vehicle's longitudinal axis; its longitudinal position is exactly half the distance from the geometric center; therefore, the aforementioned physical center point can be uniquely equivalent to the two points P1 and P2 given in the formula.

[0042] Step S2: Establish a local coordinate system F with the global reference path and the lateral deviation direction as the coordinate axes;

[0043] The formula is as follows:

[0044] ;

[0045] in, s This is the arc length distance from the projection point of the AGV along the global reference path to the starting point; d This represents the vertical distance from the AGV to the global reference path, indicating the degree of deviation of the current AGV in the direction perpendicular to the global reference path.

[0046] Step S3: Calculate the lateral offsets of positions P1 and P2 from the global reference path. d 1 and d 2 It is used to calculate and issue control signals, increasing the efficiency of the control algorithm while reducing the burden caused by large amounts of computation under high-frequency operations. Specifically, it includes the following:

[0047] Step S3.1: Traverse all discrete trajectory points on the global reference path Calculate the coordinates of each discrete trajectory point relative to the current positions P1 and P2. x 1， y 1) ( x 2， y 2) The Euclidean distance between them is given by the following formula:

[0048] ;

[0049] in, i For discrete trajectory points The number of a certain trajectory point; d 1i and d 2i These represent the points from P1 and P2 to the offline trajectory points, respectively. The distance;

[0050] Step S3.2: Select the first for P1 n discrete trajectory points Let P2 be the point with the minimum distance; select the first point for P2. m discrete trajectory points The point with the minimum distance;

[0051] Step S3.3: Calculate the longitudinal position s1 of P1 and the longitudinal position s2 of P2; where s1 is the starting point of the global reference path. To the n discrete trajectory points The cumulative arc length of all paths between them; s2 is the starting point of the global reference path. To the m discrete trajectory points The cumulative arc length of all paths between them; the formula is as follows:

[0052] ;

[0053] in, x k , y k For the first k The coordinates of a discrete trajectory point in the Cartesian coordinate system x k-1 , y k-1 For the first k -1 coordinates of discrete trajectory points in Cartesian coordinate system k The value is an index that increases progressively from 1 to n and m;

[0054] Step S3.4: Using discrete trajectory points and Using the reference point, calculate the normal direction N. r1 Normal direction N r2 ;

[0055] For P1 passing through the reference point Compared with its previous discrete trajectory point Construct the tangent direction vector T r1 The formula is as follows:

[0056] ;

[0057] Among them, T r1 The corresponding normal direction N r1 Rotate the tangent vector by 90°;

[0058] ;

[0059] Among them, T y1 Tangent vector T r1 In the Cartesian coordinate system y Projected components on the axis, Tx1 Tangent vector T r1 In the Cartesian coordinate system x Projected components on the axis;

[0060] For point P2, it passes through the reference point. Compared with its previous discrete trajectory point Construct the tangent direction vector T r2 The formula is as follows:

[0061] ;

[0062] Among them, T r2 The corresponding normal direction N r2 Rotate the tangent vector by 90°:

[0063] ;

[0064] Among them, T y2 Tangent vector T r2 In the Cartesian coordinate system y Projected components on the axis, T x2 Tangent vector T r2 The projection components on the x-axis of the Cartesian coordinate system;

[0065] Step S3.5: Based on the normal direction N r1 and normal direction N r2 Point the coordinates of positions P1 and P2 toward the reference point. and By calculating the vector difference, we obtain the vector differences ΔU1 and ΔU2, as shown in the following formula:

[0066] ;

[0067] Calculate the lateral offsets of P1 and P2 using vector projection. d 1 、d 2 The formula is as follows:

[0068] ;

[0069] .

[0070] Step S4: Based on the lateral offset d 1 、d 2 The magnetic navigation deviation data, namely the first drive overall center deviation Δ1 and the second drive overall center deviation Δ2, are obtained by conversion; specifically, the following are included:

[0071] Step S4.1: Set the coordinates of P1 (x 1， y 1) and the coordinates of P2 ( x 2， y 2) Perform coordinate system transformations to obtain P1(s) 1， d 1) and P2(s 2， d 2);

[0072] Step S4.2: Based on the lateral offset d 1 、d 2 The magnetic navigation deviation data, namely the first drive overall center deviation Δ1 and the second drive overall center deviation Δ2, are obtained by conversion, as shown in the following formula:

[0073] ;

[0074] Note: The aforementioned magnetic navigation deviations Δ1 and Δ2 are semantically equivalent to the lateral deviation signals of traditional physical magnetic sensors (e.g., the deviation value in mm returned by the magnetic sensor) in terms of controller interface. Therefore, they can directly replace the role of physical magnetic sensors in the control system without relying on ground magnetic strips or magnetic field measurements. In the above formulas, K 0 represents the proportionality coefficient.

[0075] In addition, to ensure the stability of the data conversion process, prevent abnormal values ​​caused by laser positioning fluctuations from causing control oscillations, and consider the actual maximum offset range of the AGV, the maximum deviation is calculated to be Δ. max and Δ min The calculated data is processed for deviation and a limiting mechanism is added.

[0076] Step S4 also includes a step of applying a limiting mechanism to the obtained first overall drive center deviation Δ1 and second overall drive center deviation Δ2, as shown in the following formula:

[0077] ;

[0078] ;

[0079] Where, Δ max For the maximum deviation and Δ min This represents the minimum deviation.

[0080] Step S5: Calculate the steering angular velocity control signal of the first drive assembly and the steering angular velocity control signal of the second drive assembly based on the center deviation Δ1 of the first drive assembly and the center deviation Δ2 of the second drive assembly.

[0081] The steering angular velocity control signal of the first drive unit is calculated using the following formula. Steering angular velocity control signal of the second drive system ;

[0082] ;

[0083] ;

[0084] A dual-point collaborative control strategy is adopted, using P1 and P2 as control input points to constrain the deviation of the vehicle's front and rear drive units, reduce tail sway, and improve trajectory tracking accuracy. Among these, These are all control parameters and need to be calibrated according to the dynamic characteristics of the AGV.

[0085] Step S6: Calculate the speed control signal;

[0086] In step S6, firstly, the kinematic center position P of the AGV is solved. x , y Distance P from the AGV's destination goal ( x goal , y goal ) length Then, according to Solving for the speed control signal specifically includes the following:

[0087] Step S6.1: Set P( x , y ) and P goal ( x goal , y goal Perform coordinate system transformation, P( x , y The transformed coordinates are ( s 0,d i ), P goal The transformed coordinates are (s goal ,0);

[0088] Step S6.2: According to s 0 and s goal The kinematic center of the AGV is known to be P( x , y Distance P from the AGV's destination goal ( x goal , y goal ) length for:

[0089] ;

[0090] Step S6.3: According to Using solution speed v ;

[0091] ;

[0092] This also includes the The specific steps for performing speed limiting are as follows:

[0093] ;

[0094] Among them, v max and v min These are the maximum and minimum speeds controlled by the AGV, respectively.

[0095] In step S6, P goal ( x goal , y goal The steps for coordinate system transformation are as follows:

[0096] Step S61.1: For P goal ( x goal , y goal It must be on the global reference path and has no offset, so the converted... d goal =0;

[0097] Step S61.2: Calculate P goal ( x goal , y goal The longitudinal position s goal The details are as follows:

[0098] P goal If is the last discrete trajectory point on the global reference path, then s goal Value is the starting point of the global reference path To P goal The cumulative arc length of all paths between them is calculated using the following formula:

[0099] ;

[0100] Where max is the number of the last discrete trajectory point on the global reference path;

[0101] Step S61.3: P goal The transformed coordinates are (sgoal ,0).

[0102] In step S6, P( x , y The steps for coordinate system transformation are as follows:

[0103] Step S62.1: Calculate P( x , y Lateral offset d from the global reference path i Specifically, iterate through all discrete trajectory points on the global reference path. Calculate the relationship between each discrete trajectory point and the current P( x , y The Euclidean distance between them is given by the following formula:

[0104] ;

[0105] Step S62.2: For P( x , y Select d i The smallest point Calculate P( x , y The vertical position s0 of ) is as follows:

[0106] Calculate the global reference path start point arrive The cumulative arc length of all paths between them, with the longitudinal position of P denoted as s0, is calculated as shown in the following formula:

[0107] .

[0108] Step S7: Adjust the front and rear positions and attitudes of the AGV according to the steering angular velocity control signal of the first drive system and the steering angular velocity control signal and speed control signal of the second drive system.

[0109] It should be noted that the control method in this application is a multi-drive collaborative control strategy. A coordinate system is established using the automatic navigation route and lateral deviation direction as coordinate axes. Coordinate transformation is achieved by calculating the AGV's coordinates in the Cartesian coordinate system and the automatic navigation route. Navigation and control are then performed within this coordinate system. The lateral offset distance is obtained by calculating the positions and coordinates of the front and rear axles separately. Based on the lateral offset, the drive angles and wheel speeds of the front and rear drive groups are calculated separately. Compared to traditional control models and traditional navigation control methods, this approach achieves higher control accuracy and better control performance, increasing control flexibility.

[0110] Test example:

[0111] The AGV in the test case is configured as follows, and the geometric parameters of the AGV body are shown in Table 1 below, which are used for mathematical derivation and performance analysis.

[0112] Table 1

[0113] parameter symbol numerical values Overall length of the vehicle body 7.0m Overall width of the vehicle body W 3.0m Distance from front and rear drive center to center A <![CDATA[L f ,L r ]]> 2.705m Front and rear drive center y-offset <![CDATA[±Y d ]]> ±0.7975m Drive size — 0.73×0.73×0.14m

[0114] Note: The origin of the coordinate system is taken as the geometric center of the vehicle body (with the front and rear drive centers equidistant).

[0115] The drive position coordinate parameters are shown in Table 2 below:

[0116] Table 2

[0117] The variables are defined in Table 3 below:

[0118] Table 3

[0119] symbol definition <![CDATA[ P 1 ，P 2]]> Front and rear drive overall center point <![CDATA[ d 1 ，d 2]]> <![CDATA[ P 1 ，P Lateral deviation from the reference path 2 Δ1, Δ2 <![CDATA[Based on d 1 ，d "magnetic navigation deviation amount similar to 2"]]> θ AGV's actual heading angle (yaw angle) <![CDATA[θ r ]]> Reference path heading angle <![CDATA[ ψ= θ - θ r ]]> AGV relative path attitude deviation <![CDATA[ω f Oh, oh r ]]> Steering angular velocity control of the front and rear drive systems AGV forward speed <![CDATA[e y ]]> Lateral deviation of AGV geometric center <![CDATA[e ψ ]]> Attitude deviation = Len Center distance between front and rear wheels (5.41m)

[0120] Based on the above configuration, the effects of using traditional Ackermann tail swing versus the front and rear separate control solution of this invention are compared as follows:

[0121] (1) The dynamic approximation of the transverse error using the traditional Ackermann model is:

[0122] ;

[0123] Of these, only the front wheel steering angle ;

[0124] When AGV exhibits slight posture deviation hour:

[0125] The rear axle will generate additional tailswing distance (geometric amplification):

[0126] ;

[0127] Substitution L =5.4m:

[0128] 1° deviation: ;

[0129] 2° deviation: ;

[0130] 3° deviation: ;

[0131] With just a 3° yaw, the rear end will swing out 28.3cm laterally, nearly 10% of the vehicle's width, which is extremely dangerous in narrow passages.

[0132] (2) The solution of this invention is as follows: the rear drive is combined with cooperative control to form negative feedback to suppress tail sway, and independent steering of front and rear dual drive is adopted, as follows:

[0133] ;

[0134] Since the rear axle steering also participates in the correction, the dynamic attitude deviation becomes:

[0135] ;

[0136] in: (Rear wheel) and The opposite direction generates a "reverse torque" that suppresses the tailswing, thus the effective model for tailswing offset becomes:

[0137] ;

[0138] in This is the rear axle coordination correction coefficient (determined by the control law).

[0139] Under typical AGV parameters, based on theoretical calculations and controller bandwidth derivation:

[0140] ;

[0141] Substituting into the tail swing deviation correction model, we obtain the following formula:

[0142] ;

[0143] Pick .

[0144] The calculation results are shown in Table 4:

[0145] Table 4

[0146] Yaw angle Ackerman tail swing (cm) The tail swing (cm) of this invention Tail swing reduction ratio 1° 9.46cm 3.78cm Reduce by 60% 2° 18.9cm 7.55cm Reduce by 60% 3° 28.3cm 11.3cm Reduce by 60%

[0147] According to the test results, the tail swing of the present invention is reduced by about 60% compared with the traditional method, which is more effective.

[0148] Finally, it should be noted that the above embodiments are merely illustrative and explanatory of the present invention, and are not intended to limit the present invention to the scope of the described embodiments. Furthermore, those skilled in the art will understand that the present invention is not limited to the above embodiments, and many more variations and modifications can be made based on the teachings of the present invention, all of which fall within the scope of protection claimed by the present invention.

Claims

1. A navigation and control method for a four-wheel drive differential-driven heavy-duty AGV, characterized in that, The method does not rely on physical magnetic strips and magnetic sensors. It treats the front and rear drives of the AGV as two separate units, which are controlled independently. The path offset of each unit is calculated using laser SLAM positioning, and then their respective steering and speed are controlled independently. The method includes the following steps: Step S1: Consider the two sets of drives on the front side of the AGV as the first drive unit and the two sets of drives on the rear side as the second drive unit. Calculate the coordinates of the center position P1 of the first drive unit. x 1 y 1) Coordinates of the overall center position P2 of the second drive ( x 2 y 2); Step S2: Establish a local coordinate system F with the global reference path and the lateral deviation direction as the coordinate axes; Step S3: Calculate the lateral offsets of positions P1 and P2 from the global reference path. d 1 and d 2 ; Step S4: Based on the lateral offset d 1 、d 2 The magnetic navigation deviation data, namely the first drive overall center deviation Δ1 and the second drive overall center deviation Δ2, are obtained respectively. Step S5: Calculate the steering angular velocity control signal of the first drive assembly and the steering angular velocity control signal of the second drive assembly based on the center deviation Δ1 of the first drive assembly and the center deviation Δ2 of the second drive assembly. Step S6: Calculate the speed control signal; Step S7: Adjust the front and rear positions and attitudes of the AGV according to the steering angular velocity control signal of the first drive system and the steering angular velocity control signal and speed control signal of the second drive system.

2. The navigation and control method for a four-wheel drive differential-driven heavy-duty AGV according to claim 1, characterized in that, In step S2: A local coordinate system F is established with the global reference path and the lateral deviation direction as the coordinate axes, as shown in the following formula: ； in, s This is the arc length distance from the projection point of the AGV along the global reference path to the starting point; d This represents the vertical distance from the AGV to the global reference path, indicating the degree of deviation of the current AGV in the direction perpendicular to the global reference path.

3. The navigation and control method for a four-wheel drive differential-driven heavy-duty AGV according to claim 2, characterized in that, Step S3 specifically includes the following: Step S3.1: Traverse all discrete trajectory points on the global reference path Calculate the coordinates of each discrete trajectory point relative to the current positions P1 and P2. x 1 y 1) ( x 2 y 2) The Euclidean distance between them is given by the following formula: ； ； in, i For discrete trajectory points The number of a certain trajectory point; d 1i and d 2i These represent the points from P1 and P2 to the offline trajectory points, respectively. The distance; Step S3.2: Select the first for P1 n discrete trajectory points Let P2 be the point with the minimum distance; select the first point for P2. m discrete trajectory points The point with the minimum distance; Step S3.3: Calculate the longitudinal position s1 of P1 and the longitudinal position s2 of P2; where s1 is the starting point of the global reference path. To the n discrete trajectory points The cumulative arc length of all paths between them; s2 is the starting point of the global reference path. To the m discrete trajectory points The cumulative arc length of all paths between them; the formula is as follows: ； ； in, x k , y k For the first k The coordinates of a discrete trajectory point in the Cartesian coordinate system x k-1 , y k-1 For the first k -1 coordinates of discrete trajectory points in Cartesian coordinate system k The value is an index that increases progressively from 1 to n and m; Step S3.4: Using discrete trajectory points and Using the reference point, calculate the normal direction N. r1 Normal direction N r2 ; For P1 passing through the reference point Compared with its previous discrete trajectory point Construct the tangent direction vector T r1 The formula is as follows: ； Among them, T r1 The corresponding normal direction N r1 Rotate the tangent vector by 90°; ； Among them, T y1 Tangent vector T r1 In the Cartesian coordinate system y Projected components on the axis, T x1 Tangential direction Quantity T r1 In the Cartesian coordinate system x Projected components on the axis; For point P2, it passes through the reference point. Compared with its previous discrete trajectory point Construct the tangent direction vector T r2 The formula is as follows: ； Among them, T r2 The corresponding normal direction N r2 Rotate the tangent vector by 90°: ； Among them, T y2 Tangent vector T r2 In the Cartesian coordinate system y Projected components on the axis, T x2 Tangent vector T r2 The projection components on the x-axis of the Cartesian coordinate system; Step S3.5: Based on the normal direction N r1 and normal direction N r2 Point the coordinates of positions P1 and P2 toward the reference point. and By calculating the vector difference, we obtain the vector differences ΔU1 and ΔU2, as shown in the following formula: ； ； Calculate the lateral offsets of P1 and P2 using vector projection. d 1 、d 2 The formula is as follows: ； 。 4. The navigation and control method for a four-wheel drive differential-driven heavy-duty AGV according to claim 3, characterized in that, Step S4 specifically includes the following: Step S4.1: Set the coordinates of P1 ( x 1 y 1) and the coordinates of P2 ( x 2 y 2) Perform coordinate system transformations to obtain P1(s1) respectively. d 1) and P2(s2) d 2); Step S4.2: Based on the lateral offset d 1 、d 2 The magnetic navigation deviation data, namely the first drive overall center deviation Δ1 and the second drive overall center deviation Δ2, are obtained by conversion, as shown in the following formula: ； ； in, K 0 represents the proportionality coefficient.

5. The navigation and control method for a four-wheel drive differential-driven heavy-duty AGV according to claim 4, characterized in that, Step S4 also includes a step of applying a limiting mechanism to the obtained first overall drive center deviation Δ1 and second overall drive center deviation Δ2, as shown in the following formula: ； ； Where, Δ max For the maximum deviation and Δ min This represents the minimum deviation.

6. The navigation and control method for a four-wheel drive differential-driven heavy-duty AGV according to claim 5, characterized in that, In step S5, the steering angular velocity control signal of the first drive system is calculated using the following formula. Steering angular velocity control signal of the second drive system ; ； ； in, These are all control parameters and need to be calibrated according to the dynamic characteristics of the AGV.

7. The navigation and control method for a four-wheel drive differential-driven heavy-duty AGV according to claim 6, characterized in that, In step S6, firstly, the kinematic center position P of the AGV is solved. x , y Distance P from the AGV's destination goal ( x goal , y goal ) length 3. Then, according to 3. Solving for the speed control signal, specifically including the following: Step S6.1: Set P( x , y ) and P goal ( x goal , y goal Perform coordinate system transformation, P( x , y The transformed coordinates are ( s 0,d i ), P goal The transformed coordinates are (s goal ,0); Step S6.2: According to s 0 and s goal The kinematic center of the AGV is known to be P( x , y Distance P from the AGV's destination goal ( x goal , y goal ) length 3 is: ； Step S6.3: According to 3. Using solution speed v ; 。 8. The navigation and control method for a four-wheel drive differential-driven heavy-duty AGV according to claim 7, characterized in that, In step S6, P goal ( x goal , y goal The steps for coordinate system transformation are as follows: Step S61.1: For P goal ( x goal , y goal It must be on the global reference path and has no offset, so the converted... d goal =0; Step S61.2: Calculate P goal ( x goal , y goal The longitudinal position s goal The details are as follows: P goal If is the last discrete trajectory point on the global reference path, then s goal Value is the starting point of the global reference path To P goal The cumulative arc length of all paths between them is calculated using the following formula: ； Where max is the number of the last discrete trajectory point on the global reference path; Step S61.3: P goal The transformed coordinates are (s) goal ,0).

9. A navigation and control method for a four-wheel drive differential-driven heavy-duty AGV according to claim 8, characterized in that, In step S6, P( x , y The steps for coordinate system transformation are as follows: Step S62.1: Calculate P( x , y Lateral offset d from the global reference path i Specifically, iterate through all discrete trajectory points on the global reference path. Calculate the relationship between each discrete trajectory point and the current P( x , y The Euclidean distance between them is given by the following formula: ； Step S62.2: For P( x , y Select d i The smallest point b Calculate P( x , y The vertical position s0 of ) is as follows: Calculate the global reference path start point to b The cumulative arc length of all paths between them, with the longitudinal position of P denoted as s0, is calculated as shown in the following formula: 。 10. A navigation and control method for a four-wheel drive differential-driven heavy-duty AGV according to claim 9, characterized in that, Step S6.3 also includes... The specific steps for performing speed limiting are as follows: ； Among them, v max and v min These are the maximum and minimum speeds controlled by the AGV, respectively.

Images