AGV (Automatic Guided Vehicle) task intelligent distribution method and system

Abstract

The invention relates to an AGV (Automatic Guided Vehicle) task intelligent distribution method and system. The AGV task intelligent distribution method is characterized in that the motion performance of each AGV is evaluated by collecting control parameters and motion parameters in the task execution process of each AGV; obtaining standby AGVs in a certain range near the carrying starting point, and reading the maximum load, the current electric quantity and the motion performance parameters of each standby AGV; according to the load ratio of each standby AGV and the ground wear degrees of the three sections, the ground slip probability of each AGV when the AGV passes through the three sections is obtained; and calculating the task adaptation degree of each AGV according to the weight of the goods, the maximum load, the load ratio, the current electric quantity and the motion performance parameters of each standby AGV, and the ground slip probabilities of the AGV passing through the three sections of the currently planned carrying path. According to the method, the timeliness and accuracy of cargo carrying can be improved, the cargo carrying efficiency is improved, and fine management of the workshop operation process is facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control, and particularly relates to a method and system for intelligent task distribution of AGV vehicles. Background Art

[0002] AGV, that is, an automatic guided vehicle, as one of the core devices of intelligent logistics, has witnessed an explosive growth in recent years. And AGVs that are mainly intelligent and can navigate autonomously have also been widely used in the field of intelligent logistics warehousing.

[0003] In actual scenarios, multiple AGVs may need to run simultaneously, which makes it particularly important whether the tasks assigned to each AGV are reasonable. Otherwise, it will cause problems such as too low operating efficiency and resource waste.

[0004] In the prior art, when assigning tasks to AGVs, generally only factors such as the maximum load, current battery level, and distance from the handling path of the AGV are considered.

[0005] In fact, the motion performance of an AGV may decline after long-term operation. When the motion performance declines, the timeliness and accuracy of its acceleration, turning, and deceleration actions during the handling process may be adversely affected, which will affect the timeliness and accuracy of goods handling. If the conventional task assignment strategy is adopted without considering the motion performance, it is very likely that an AGV with poor motion performance will be assigned to execute the task, which may have a certain impact on the efficiency of handling goods and is not conducive to the refined management of the workshop operation process and needs to be improved. Summary of the Invention

[0006] Based on the above description, an object of the present invention is to provide a method and system for intelligent task distribution of AGV vehicles, which comprehensively consider factors such as the status, motion performance, and distance of each AGV vehicle when distributing tasks, and automatically select the best AGV vehicle to execute the next goods handling task.

[0007] The technical solution for the present invention to solve the above technical problems is as follows:

[0008] An intelligent task distribution method for AGV vehicles: During the process of each AGV vehicle executing tasks, its motion performance is evaluated by collecting control parameters and motion parameters. The motion performance includes acceleration performance, steering performance, and braking performance; The handling path is automatically planned according to the starting point and ending point of the goods; Obtain the acceleration section, steering section, and deceleration section on the handling path, and evaluate the ground wear degree of the three sections respectively; Obtain the standby AGV vehicles within a certain range near the handling starting point, and read the maximum load, current battery level, and motion performance parameters of each standby AGV vehicle; According to the load ratio of each standby AGV and the ground wear degree of the above three sections, obtain the ground slip probability of each AGV vehicle when passing through the three sections; Calculate the task suitability of each AGV vehicle based on the weight of the goods, the maximum load, load ratio, current battery level, motion performance parameters of each standby AGV vehicle, and the ground slip probability of the three sections of the current planned handling path passed by the AGV vehicle; Sort the task suitability of each AGV vehicle, and select the AGV vehicle with the highest task suitability to execute this handling task.

[0009] As a preferred solution: Evaluating the acceleration performance of the AGV vehicle is to collect the rising value of the drive current of the drive wheel motor within a fixed time period during the acceleration stage, and convert it into the rising value of the drive torque of the drive wheel accordingly; At the same time, detect the rising value of the driving speed of the AGV vehicle during this fixed time period, and then combine the current load to obtain the theoretical rising value of the driving speed. Finally, calculate the ratio of the actual rising value of the driving speed to the theoretical rising value of the driving speed, and evaluate the acceleration performance of the AGV vehicle based on this.

[0010] As a preferred solution: Evaluating the braking performance of the AGV vehicle is to collect the rising value of the braking current of the brake within a fixed time period during the deceleration stage, and convert it into the rising value of the braking torque of the brake accordingly; At the same time, detect the decreasing value of the driving speed of the AGV vehicle during this fixed time period, and then combine the current load to obtain the theoretical decreasing value of the driving speed. Finally, calculate the ratio of the actual decreasing value of the driving speed to the theoretical decreasing value of the driving speed, and evaluate the braking performance of the AGV vehicle based on this.

[0011] As a preferred solution: Evaluating the steering performance of the AGV vehicle is to start collecting the angular velocity signal of the gyroscope sensor at the same time when the motion controller sends a steering command to the steering servo to obtain the time point when the heading angle changes. Calculate the steering delay time based on the time point when the heading angle changes and the time point when the steering command is sent, and compare the calculated delay time with the preset allowable delay time to evaluate the steering performance of the AGV vehicle.

[0012] As a preferred solution: To evaluate the ground wear degree, first count the cumulative number of AGV vehicles passing through the acceleration section, steering section, and deceleration section and the cumulative load of the passing vehicles, and then calibrate the ground wear degree of the standard grid through experiments based on the cumulative number of passing vehicles and the cumulative load of the passing vehicles.

[0013] As a preferred solution: To obtain the ground slip probability when each AGV vehicle passes through the three sections, the method of controlling variables is used, that is, under various different load ratios, control the AGV to pass through the ground with various different wear degrees multiple times, and then detect and count the number of slips when it passes through the ground with different wear degrees. Based on this, the slip probability of various types of AGV vehicles when passing through the ground with different wear degrees under different load ratios can be obtained.

[0014] An AGV vehicle task intelligent distribution system for executing the above method, the system includes:

[0015] A data acquisition module, which is used to acquire AGV vehicle identification information, position information, status data, and motion parameters;

[0016] A storage module, which is used to store data and information;

[0017] A path planning module, which is used to automatically plan and generate a handling path according to the handling starting point and ending point of the goods;

[0018] A section positioning module, which is used to identify and position the acceleration section, steering section, and deceleration section in the generated handling path;

[0019] A slip evaluation module, which is used to evaluate the ground slip probability of each AGV passing through the acceleration section, steering section, and deceleration section;

[0020] A performance evaluation module, which is used to evaluate the motion performance of the standby AGV vehicles within a certain range near the handling starting point, including acceleration performance, steering performance, and braking performance;

[0021] A matching module, which calculates the task suitability of each AGV vehicle based on the weight of the goods and the maximum load, current power, motion performance parameters of each standby AGV vehicle, and the ground slip probability of the AGV passing through the three sections of the currently planned handling path; sort the task suitability of each AGV vehicle, and select the AGV vehicle with the highest task suitability, that is, the selected vehicle;

[0022] An instruction sending module, which is used to send an instruction to the selected vehicle to execute the handling task.

[0023] Compared with the prior art, the technical solution of the present application has the following beneficial technical effects: When distributing tasks, the present invention comprehensively considers factors such as the status, motion performance, distance of each AGV cart, and the skidding probability when each AGV cart passes through the acceleration section, steering section, and deceleration section on the semicircular path, and automatically matches and selects the best AGV cart to execute the next cargo handling task. This method can improve the timeliness and accuracy of cargo handling, improve the cargo handling efficiency, and is conducive to the refined management of the workshop operation process. Description of the Drawings

[0024] Figure 1 It is a schematic flowchart of the method in the embodiment;

[0025] Figure 2 It is a schematic diagram of the system in the second embodiment. Detailed Embodiments

[0026] Embodiment 1:

[0027] An intelligent task distribution method for AGV carts, and the method is as follows:

[0028] S1. During the process of each AGV cart executing tasks, evaluate its motion performance by collecting control parameters and motion parameters. The motion performance includes acceleration performance, steering performance, and braking performance.

[0029] When the AGV cart is first put into use in the workshop, its mechanical state and electrical state are in the best state, and its motion performance such as acceleration, steering, and braking is the best. However, as the usage time increases and the number of tasks executed increases, electrical aging and mechanical wear will inevitably occur, resulting in a decline in its motion performance.

[0030] For AGV carts of the same model but with different motion performances, when carrying the same weight of goods, the acceleration time required from starting to reaching the maximum running speed, the number of times of correcting the required steering angle during steering, and the braking time required to decelerate to a standstill are different. The accumulation of these factors may lead to obvious differences in the final cargo handling efficiency.

[0031] In order to improve the cargo handling efficiency, it is necessary to evaluate the motion performance of the AGV cart before it executes the next handling task. Here, the evaluation of the motion performance is based on the control parameters and motion parameters of the AGV cart during the previous task execution process. The evaluation result is used as the basis for whether to select the AGV cart for the next task execution.

[0032] In this embodiment, the acceleration performance of the AGV cart is evaluated by the following method:

[0033] The acceleration stage mentioned here refers to the time period during which the speed of the AGV vehicle accelerates from zero to the default maximum traveling speed. During the acceleration stage, it gradually moves away from the handling starting point. Generally speaking, for AGV vehicles of the same model, the rising curve of the drive current of the drive motor during the acceleration stage is fixed (it can be made fixed through program control). The rising curve of the drive current mentioned here refers to the drive current characteristic curve plotted with time as the abscissa and the drive current of the drive motor as the ordinate.

[0034] Therefore, the rising value of the drive current collected within a fixed time period during the acceleration stage is also fixed.

[0035] During the acceleration stage, by collecting the rising value of the drive current of the drive wheel motor within a fixed time period (for example, within the time period from 3s to 5s after starting, or it can be other time periods, as long as the collection time period is within the acceleration stage of the vehicle), the rising value of the drive wheel torque is calculated accordingly. For example, the drive current collected at 3s after starting is iq1, and the drive current collected at 5s after starting is iq2, then the drive current rising value Δqi = iq2 - iq1; using the motor torque calculation formula T = K * i (where T represents the torque of the motor, K represents the magnetic torque constant of the motor, and i represents the current symbol), the rising value of the drive wheel torque ΔT = K * Δqi is calculated. Since the current rising curve is fixed and the sampling time period is also fixed, the calculated torque rising value ΔT is also fixed.

[0036] Therefore, for AGV vehicles of the same model, a large number of experiments can be carried out in advance to use them to handle goods of different weights, and the speed rising value within 3s - 5s during the acceleration stage under different load conditions can be detected, that is, the calibration of the theoretical traveling speed rising value is completed.

[0037] During the actual task execution process, in addition to the above process of collecting the drive current rising value and conversion, on the premise that the drive current rising value remains unchanged (that is, when the electrical performance is normal), it is also necessary to detect the actual speed rising value of the AGV vehicle within this fixed time period (3s - 5s) during the acceleration stage; combined with the current load (the load can be obtained by reading the goods information), the previously calibrated theoretical speed rising value is obtained, and finally the ratio of the actual speed rising value to the theoretical speed rising value is calculated, and based on this, the acceleration performance of the AGV vehicle is evaluated.

[0038] Assume that the default acceleration performance is assigned a value of 1. If the ratio of the detected and calculated actual speed rising value to the theoretical speed rising value is 09:1, it is considered that the actual acceleration performance of this AGV vehicle is 0.9; if the ratio is 08:1, it is considered that the actual acceleration performance of this AGV vehicle is 0.8, and so on.

[0039] In this embodiment, the method for evaluating the braking performance of the AGV vehicle is as follows:

[0040] In this embodiment, the deceleration stage refers to the time period during which the AGV vehicle decelerates from the default maximum traveling speed to zero.

[0041] Generally speaking, for AGV vehicles of the same model, the rising curve of the braking current of the brake (usually an electromagnetic brake) during the deceleration stage is fixed (it can be made fixed through program control). The rising curve of the braking current mentioned here refers to the braking current characteristic curve plotted with time as the abscissa and the braking current of the brake as the ordinate.

[0042] Therefore, the rising value of the braking current collected within a fixed time period (for example, 3s - 5s) during the deceleration stage is also fixed.

[0043] During the deceleration stage, collect the rising value of the braking current of the brake within a fixed time period (3s - 5s), and convert it into the rising value of the braking torque of the brake accordingly.

[0044] For example, if the driving current collected at 3s after starting braking is iz1 and the driving current collected at 5s after starting braking is iz2, then the rising value of the driving current Δzi = iz2 - iz1; using the braking torque calculation formula M = F * d (where M represents the braking torque of the brake, F represents the braking force of the brake, and d represents the distance from the braking force application point) and the braking force calculation formula F = B * I * L (where F represents the braking force, B represents the magnetic induction intensity, I represents the braking current, and L represents the conductor length), we can calculate M = B * I * L * d. Then, substituting the rising value of the braking current Δzi into the above formula, we can obtain the rising value of the braking torque ΔM = B * Δzi * L * d.

[0045] Since the rising curve of the braking current is fixed and the sampling time period is also fixed, the calculated rising value of the braking torque ΔM is also fixed.

[0046] Therefore, for AGV vehicles of the same model, a large number of experiments can be carried out in advance to have them carry goods of different weights and detect the speed drop value within 3s - 5s during the deceleration stage under different load conditions, that is, to complete the calibration of the theoretical traveling speed drop value.

[0047] In the actual process of performing tasks, in addition to the above process of collecting the rising value of the braking current and conversion, on the premise that the rising value of the braking current remains unchanged (i.e., under normal electrical performance), it is also necessary to detect the actual speed reduction value of the AGV during the fixed time period (3s - 5s) in the deceleration stage; then combine the current load (the load can be obtained by reading the cargo information) to obtain the calibrated theoretical speed reduction value, and finally calculate the ratio of the actual speed reduction value to the theoretical speed reduction value, and evaluate the braking performance of the AGV accordingly.

[0048] Assume that the default braking performance is assigned a value of 1. If the ratio of the detected and calculated actual speed reduction value to the theoretical speed reduction value is 09:1, it is considered that the actual braking performance of the AGV is 0.9; if the ratio is 08:1, it is considered that the actual braking performance of the AGV is 0.8, and so on.

[0049] In this embodiment, the method for evaluating the steering performance of the AGV is as follows: While the motion controller sends a steering command to the steering servo, start collecting the angular velocity signal of the gyroscope sensor to obtain the time point when the heading angle changes. Calculate the steering delay time based on the time point when the heading angle changes and the time point when the steering command is issued, and calculate the ratio of the steering delay time to the preset reference delay time, and evaluate the steering performance of the AGV accordingly.

[0050] Assume that the default steering performance is assigned a value of 1. If the ratio of the steering delay time to the reference delay time is 09:1, it is considered that the actual steering performance of the AGV is 0.9; if the ratio is 08:1, it is considered that the actual steering performance of the AGV is 0.8, and so on.

[0051] S2. Automatically plan the handling path according to the starting point and ending point of the cargo; obtain the acceleration section, steering section, and deceleration section on the handling path, and evaluate the ground wear degree of the three sections respectively.

[0052] In this embodiment, the driving area in the workshop can be divided into multiple grids and numbered. After the handling path is automatically planned, identify the starting position, steering position, and ending position of the handling path, and then determine the grids where these positions are located, that is, the target grids. Define the grid area where the starting position is located as the acceleration section, the grid area where the steering position is located as the steering section, and the grid area where the ending position is located as the deceleration section.

[0053] It should be noted that: when there are more than two turning points on the handling path, in order to simplify data processing and calculation amount, in this embodiment, one of the turning points is randomly selected to represent the steering position of the AGV.

[0054] Obtain the cumulative number of AGV cars passing through the acceleration section, turning section, and deceleration section, and evaluate the ground wear degree of each section accordingly. Here, the definition of "cumulative" means starting from the time when the workshop is put into use, or starting from the time when the workshop floor is renovated and put into use again.

[0055] Generally, the more cumulative AGV units passing through a grid area, the more serious the ground wear of that grid area. In this embodiment, a large number of tests can be carried out in advance, and the ground wear degree of the standard grid can be calibrated according to the cumulative number of passing vehicles and the cumulative load of the passing vehicles.

[0056] After determining the acceleration section, turning section, and deceleration section and obtaining the cumulative number of passing vehicles and the cumulative load of the passing vehicles corresponding to the grids in each section, the ground wear degree corresponding to each section can be obtained.

[0057] S3. Obtain the standby AGV cars within a certain range near the handling starting point, and read the maximum load, current power, and motion performance parameters of each standby AGV car.

[0058] After determining the handling starting point, read the position information, power information, identification information, maximum load, and motion performance parameters of each idle and standby AGV car. According to the distance between the position of the idle AGV car and the position of the handling starting point, the maximum load of the AGV car, and the power of the AGV car, the AGV cars available for dispatch are pre-screened. For example, the AGV cars with a distance less than 50 meters from the handling starting point, a maximum load greater than the weight of this shipment, and a power greater than 20% are pre-selected as the cars available for dispatch.

[0059] S4. Obtain the ground skidding probability of each AGV car when passing through the three sections according to the type, load ratio, and the ground wear degree of the aforementioned three sections of each standby AGV.

[0060] In this embodiment, the load ratio refers to the ratio of the actual load of the AGV car to its maximum load. It is necessary to conduct tests on various types of AGV cars in advance. By the method of controlling variables, that is, under various different load ratios, control the AGV to pass through the ground with various different wear degrees multiple times, and then detect and count the number of skidding times when it passes through the ground with different wear degrees. Accordingly, the skidding probability of various types of AGV cars when passing through the ground with different wear degrees under different load ratios can be obtained.

[0061] S5. Calculate the task adaptability of each AGV car according to the weight of the goods, the maximum load, load ratio, current power, motion performance parameters of each standby AGV car, and the ground skidding probability of the three sections of the current planned handling path passed by the AGV car.

[0062] In this embodiment, a task matching degree calculation formula is preset:

[0063] W = a*U / p1 + b*H / p2 + c*Y / p3, where W represents the task matching degree value; a, b, and c are the calculation coefficients for the preset acceleration section, turning section, and deceleration section respectively; U, H, and Y are the acceleration performance parameter, turning performance parameter, and braking performance parameter of the AGV car respectively; p1, p2, and p3 are the skidding probabilities of the AGV passing through the acceleration section, turning section, and deceleration section respectively.

[0064] After obtaining the load ratio and the ground wear degrees of the three sections, the corresponding p1, p2, and p3 values can be retrieved.

[0065] S6. Sort the task adaptability degrees of each AGV car, and select the AGV car with the highest task adaptability degree to execute this handling task.

[0066] Through the previous step, the task matching degrees of each available AGV car for the current cargo handling can be calculated. Then, sort the magnitudes of the task matching degrees, and select the AGV car with the highest task matching degree to execute the task of handling this cargo.

[0067] As a preferred solution: Evaluate the ground skidding probabilities of the acceleration section, turning section, and deceleration section. First, obtain the number of AGV cars traveling daily in the area where each section is located, and then estimate the ground skidding probabilities of the three sections in combination with the cumulative working days in the workshop.

[0068] As a preferred solution: When evaluating the ground skidding probability, also detect the environmental temperature in the workshop, and correct the ground skidding probabilities of the acceleration section, turning section, and deceleration section according to the real-time environmental temperature.

[0069] As a preferred solution: When evaluating the ground skidding probability, also detect the ground humidity, and correct the ground skidding probabilities of the acceleration section, turning section, and deceleration section according to the ground humidity.

[0070] Embodiment 2:

[0071] An intelligent task distribution system for AGV cars, which is used to execute the method in Embodiment 1. This system includes:

[0072] A data acquisition module, which is used to acquire the AGV car identification information, position information, status data, and motion parameters;

[0073] A storage module, which is used to store data and information;

[0074] A path planning module, which is used to automatically plan and generate a handling path according to the starting point and ending point of the cargo;

[0075] A section positioning module, which is used to identify and locate the acceleration section, turning section and deceleration section in the generated handling path;

[0076] A skid evaluation module, which is used to evaluate the ground skid probability of each AGV passing through the acceleration section, turning section and deceleration section;

[0077] A performance evaluation module, which is used to evaluate the motion performance of the standby AGV cars within a certain range near the handling starting point, including acceleration performance, turning performance and braking performance;

[0078] A matching module, which calculates the task fitness of each AGV car according to the weight of the goods, the maximum load, the current power, the motion performance parameters of each standby AGV car, and the ground skid probability of the three sections of the currently planned handling path; sorts the task fitness of each AGV car, and screens out the AGV car with the highest task fitness, that is, the selected car;

[0079] An instruction sending module, which is used to send an instruction to the selected car to execute the handling task.

[0080] The above are only the preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent task distribution method for an AGV cart, characterized in that: During the process of each AGV carrying out tasks, its motion performance is evaluated by collecting control parameters and motion parameters. The motion performance includes acceleration performance, steering performance, and braking performance. The carrying path is automatically planned according to the starting point and ending point of the goods. The acceleration section, steering section, and deceleration section on the carrying path are obtained, and the ground wear degrees of the three sections are evaluated respectively. The standby AGV cars within a certain range near the starting point of carrying are obtained, and the maximum load, current power, and motion performance parameters of each standby AGV car are read. According to the load ratio of each standby AGV and the ground wear degrees of the above three sections, the ground slip probability of each AGV car when passing through the three sections is obtained. According to the weight of the goods, the maximum load, load ratio, current power, motion performance parameters of each standby AGV car, and the ground slip probability of this AGV car when passing through the three sections of the currently planned carrying path, the task fitness of each AGV car is calculated. The task fitness of each AGV car is sorted, and the AGV car with the highest task fitness is selected to execute this carrying task.

2. The intelligent task distribution method for the AGV cart according to claim 1, wherein: To evaluate the acceleration performance of the AGV car, during the acceleration stage, the rising value of the drive current of the drive wheel motor within a fixed time period is collected, and based on this, the rising value of the drive wheel torque is converted. At the same time, the rising value of the driving speed of the AGV car within this fixed time period is detected, and then combined with the current load to obtain the theoretical rising value of the driving speed. Finally, the ratio of the actual rising value of the driving speed to the theoretical rising value of the driving speed is calculated, and based on this, the acceleration performance of the AGV car is evaluated.

3. The intelligent task distribution method for the AGV vehicle according to claim 1, characterized in that: To evaluate the braking performance of the AGV car, during the deceleration stage, the rising value of the braking current of the brake within a fixed time period is collected, and based on this, the rising value of the brake torque is converted. At the same time, the decreasing value of the driving speed of the AGV car within this fixed time period is detected, and then combined with the current load to obtain the theoretical decreasing value of the driving speed. Finally, the ratio of the actual decreasing value of the driving speed to the theoretical decreasing value of the driving speed is calculated, and based on this, the braking performance of the AGV car is evaluated.

4. The intelligent task distribution method for the AGV cart according to claim 1, characterized in that: To evaluate the steering performance of the AGV car, while the motion controller sends a steering command to the steering servo, the angular velocity signal of the gyroscope sensor is collected to obtain the time point when the heading angle changes. According to the time point when the heading angle changes and the time point when the steering command is sent, the steering delay time is calculated. The calculated delay time is compared with the preset allowable delay time, so as to evaluate the steering performance of the AGV car.

5. The intelligent task distribution method for the AGV cart according to claim 1, characterized in that: To evaluate the ground wear degree, first, the cumulative number of AGV cars passing through and the cumulative load of the passing vehicles in the acceleration section, steering section, and deceleration section are counted. Then, through experiments, the ground wear degree of the standard grid is calibrated according to the cumulative number of passing vehicles and the cumulative load of the passing vehicles.

6. The intelligent task distribution method for the AGV vehicle according to claim 1, characterized in that: Obtaining the ground slip probability of each AGV when passing through three types of sections is achieved through the method of controlling variables. That is, under various different load ratios, the AGV is repeatedly controlled to pass through the ground with various degrees of wear, and then the number of slips when passing through the ground with different degrees of wear is detected and counted. Based on this, the slip probability of various types of AGV when passing through the ground with different degrees of wear under different load ratios can be obtained.

7. An AGV car task intelligent distribution system for performing the method described in any one of claims 1-6, characterized in that, The system includes: A data acquisition module, which is used to collect AGV identification information, position information, status data, and motion parameters; A storage module, which is used to store data and information; A path planning module, which is used to automatically plan and generate a handling path according to the starting point and ending point of the goods; A section positioning module, which is used to identify and position the acceleration section, turning section, and deceleration section in the generated handling path; A slip evaluation module, which is used to evaluate the ground slip probability of each AGV when passing through the acceleration section, turning section, and deceleration section; A performance evaluation module, which is used to evaluate the motion performance of the standby AGV within a certain range near the handling starting point, including acceleration performance, turning performance, and braking performance; A matching module, which calculates the task fitness of each AGV according to the weight of the goods, the maximum load of each standby AGV, the current battery level, the motion performance parameters, and the ground slip probability of the three sections of the currently planned handling path; sorts the task fitness of each AGV, and filters out the AGV with the highest task fitness, that is, the selected AGV; An instruction sending module, which is used to send an instruction to the selected AGV to execute the handling task.

Images