An AGV control system and control method based on QR code positioning and navigation

Abstract

The present invention discloses an AGV control system and a control method based on two-dimensional code positioning navigation, and belongs to the field of AGV control. The method optimizes the initialization step of the trolley power-on, provides a simple automatic homing program, and solves the shortcomings of the traditional AGV being in a silent state after power-on and unable to automatically and simply homing through a new homing method, achieving the purpose of automatically eliminating angle deviations at startup and simplifying the complexity of subsequent corrections; through a multi-strategy correction scheme, the problem of the singleness of the traditional correction method and the narrow application surface is solved, achieving the purpose of simplifying the correction method and effectively reducing the correction time; through multi-sensor data fusion, the problem of single data acquisition and one-sided data acquisition in the traditional correction scheme, which leads to the failure of the correction strategy, is solved, and the stability and correction accuracy of the AGV control system are effectively improved.

Description

Technical Field

[0001] The invention relates to an AGV control system and a control method thereof based on two-dimensional code positioning navigation, and belongs to the field of AGV control. Background Art

[0002] With the development of science and technology and the major changes in shopping methods, logistics sorting equipment plays an increasingly important role in the fields of manufacturing, sorting transportation and resource allocation. According to the latest development report on logistics and warehousing equipment, the coverage rate of automated equipment in most sorting warehouses in my country is still very low, and the sorting work is mainly completed by manpower, with labor costs accounting for more than 30% of the total cost.

[0003] In order to reduce labor costs and improve the space utilization of sorting warehouses, intelligent unmanned warehousing has gradually become the mainstream research direction in the field of logistics. With the deepening of research, the Automated Guided Vehicle (AGV) has begun to enter the industry's field of vision as a new type of automated logistics equipment that can reduce handling time, management and labor costs. AGV is an automated transportation equipment that uses a high-power battery on board as a power source and is equipped with non-contact guidance devices such as electromagnetics or optics. The equipment can travel along a pre-planned guidance path while loading parcels. When fully charged, AGV can work continuously for more than 4 hours at a speed of 1-2m / s. At the same time, it can store all relevant information generated during the sorting process in a real-time updated database, which can not only greatly improve the warehouse sorting efficiency, but also greatly reduce labor and management costs. With the continuous improvement of AGV functions, this equipment is also playing an increasingly important role in related fields.

[0004] In recent years, with the continuous advancement of science and technology, AGV guidance methods have also been continuously developed. There are many commonly used AGV guidance methods, such as electromagnetic guidance, magnetic tape guidance, QR code guidance, optical guidance and other fixed path guidance, inertial guidance, SLAM guidance, GPS guidance and other free path guidance. Different guidance methods can be selected according to the type, structure and application scenario of the AGV itself.

[0005] The electromagnetic guidance method is not easy to be polluted and damaged, but the driving path cannot be changed, so the electromagnetic guidance is suitable for sites where the path does not need to change frequently and the road conditions are relatively simple. Inertial guidance can only obtain the position, posture, direction and speed of the AGV itself, and cannot obtain the position information of the AGV in the map, so inertial guidance is often used in combination with other guidance methods. SLAM navigation technology has high positioning accuracy, does not require the laying of guide rails, and the line can be flexibly planned. It is a very promising navigation method. However, the cost of this technology is relatively high, and the working environment requirements are relatively harsh, so there are currently few mature products. The QR code guidance consists of components such as a code reader and a central controller. The central controller processes the code reader data to obtain the AGV position and posture information. This method is one of the common AGV guidance methods on the market and is relatively simple to implement.

[0006] Although the development of AGV based on QR code navigation has become mature, there is still room for improvement in the correction method and overall system structure.

[0007] AGV deviation control is to adjust the speed of the AGV according to the position deviation information collected during the driving process so that the AGV can maintain the correct running state. When the AGV deviates from the predetermined track, the controller can make the AGV return to the correct track within a certain period of time, that is, track tracking.

[0008] In order to solve the problem of low accuracy of AGV docking position, Zhang Yan used the location information in the Data Matrix QR code for navigation and positioning. In order to adapt to different road conditions, a two-step correction method was used to correct the position during the tracking process. Finally, the AGV path tracking was achieved by first using the two-step correction method for position correction, and then using integral separation type proportional integral differential (Proportion Integration Differentiation, PID) to achieve angle correction (Zhang Yan, Zhu Zhenwei. Research on QR Code Navigation AGV Control System [J]. Journal of Shandong Agricultural University (Natural Science Edition), 2019, 50(03): 441-444). Xia Popo obtained the deviation distance between the AGV body and the guide line based on the infrared sensor, used the dynamic deviation to realize the judgment of the forward distance, and proposed an improved pure tracking method to solve the path tracking problem of different initial states (Xia Popo, Yao Lijian, Yang Zidong, Zhang Ruifeng, Zhao Chenyan. Path tracking of greenhouse AGV transport platform based on improved pure tracking model [J]. Agricultural Mechanization Research, 2019, 41(11): 52-56+61. DOI: 10.13427 / j.cnki.njyi.2019.11.009.). Shi Wen proposed a P-proportional fuzzy control structure based on the improved immune method to correct the deviation of the AGV tracking path and reduce the tracking error (Shi Wen, Zhang Wei, Wang Yagang. P-fuzzy control of AGV path deviation correction based on immune algorithm [J]. Control Engineering, 2021, 28(05): 870-876. DOI: 10.14107 / j.cnki.kzgc.20190267.). Luo Zai used the optimal deviation conversion method to establish an AGV correction controller, which can quickly and smoothly eliminate the heading deviation (Luo Zai, Tang Yingqi, Li Dong, Wang Lanjing. Automatic guided vehicle correction method based on optimal deviation path [J]. Chinese Journal of Scientific Instruments, 2017, 38(04): 853-860. DOI: 10.19650 / j.cnki.cjsi.2017.04.009.). Li Zhao used QR code as navigation information and fuzzy control to realize AGV path correction (Li Zhao, Shu Zhibing, Yan Liang. Research on AGV visual precision positioning based on fuzzy path correction [J]. Application of Electronic Technology, 2018, 44(04): 81-85. DOI: 10.16157 / j.issn.0258-7998.173846.). Zhang Kun used fuzzy control to realize the steering control of differential AGV to improve its steering control accuracy and speed (Zhang Kun, Xu Lunhui. Research on AGV differential steering control algorithm based on fuzzy control [J]. Automation and Instrumentation, 2016, 31(10): 1-4+66. DOI: 10.19557 / j.cnki.1001-9944.2016.10.001.).

[0009] In the above-mentioned literature, the control method has a good tracking effect in the path tracking control process. However, the above-mentioned control method is mostly based on single data information for control, the correction complexity is high, and it takes too long time. It does not fully consider the instability caused by the highly complex geometric path when the car is driving at high speed. As a result, the correction accuracy and stability of the AGV control system are relatively poor. Summary of the invention

[0010] In order to effectively reduce the complexity and time consumption of correction, and improve the stability of the overall system operation, the present invention provides an AGV control system based on QR code positioning navigation and a control method thereof. The technical solution is as follows:

[0011] The first object of the present invention is to provide an AGV control system based on two-dimensional code positioning and navigation, comprising: a host computer and an AGV vehicle-mounted system, wherein the host computer comprises:

[0012] The map monitoring module is used to generate an electronic map based on the QR code label and update the vehicle's position and operating status in real time on the map;

[0013] Status information module, used to observe the status information and network connection information of the car in real time;

[0014] The command sending module is used to send action commands to the car;

[0015] The path planning module is used to search for the optimal path based on the starting point and target point of the car through the ant colony method, and finally obtain the optimal path;

[0016] The AGV vehicle-mounted system is used to realize the actual operation of the AGV; the AGV vehicle-mounted system includes: a main board and a slave board;

[0017] The main board includes: a real-time scheduling system for cyclically querying feedback information of different functional modules of the car. After receiving the feedback information, the main board packages the feedback information and uploads it to the host computer through the wireless communication module;

[0018] The slave board includes: a code reader, a digital micro-inertial navigation system and a motor drive module; after the code reader reads the current posture information, it is uploaded to the host computer, and the host computer sends a path data packet or a command data packet according to the current posture; after the slave board receives the path data packet or the command data packet, it starts to execute the command, and the motor drive module drives the motor to synchronously go straight to a specified distance or turn to a specified direction; the digital micro-inertial navigation system calculates the heading angle of the car based on the gyroscope and the accelerometer, and when the car passes through the blank area between two QR codes, it combines the feedback information of the motor encoder to infer the current posture of the car, and corrects the lateral and angular deviations that occur during operation. When the car passes through different ground QR codes, it continuously uploads posture information and continuously adjusts the posture to prevent error accumulation and ultimately cause code loss.

[0019] Optionally, the modules carried by the mainboard include: an obstacle avoidance radar module, a wireless communication module, a power module, a command transceiver module, an LCD display module and a FreeRTOS system;

[0020] The real-time scheduling system adopts the FreeRTOS system;

[0021] The feedback information includes: the power supply voltage information fed back by the power module, the obstacle approach information fed back by the obstacle avoidance radar module, the path information fed back by the slave board, and the command completion feedback information;

[0022] The LCD display module displays the battery information of the vehicle;

[0023] The command transceiver module realizes the command transceiver function between the main board and the slave board through serial port communication.

[0024] Optionally, the host computer sends the optimal path generation path data packet to the AGV vehicle-mounted system, and the path data includes a plurality of node information, and each node information includes: an address code ID, an action code, speed gear information and action parameters.

[0025] Optionally, the AGV vehicle-mounted system is built based on STM32, and the main board and the slave board are STM32 boards.

[0026] Optionally, the motor drive module adopts a differential drive mode and is developed based on the CANopen protocol.

[0027] Optionally, the host computer is built based on the JavaFX architecture.

[0028] The second object of the present invention is to provide an AGV deviation correction method based on two-dimensional code positioning and navigation, which is implemented based on the above-mentioned AGV control system based on two-dimensional code positioning and navigation, and includes:

[0029] Step 1: After the AGV is powered on, when the barcode reader reads valid QR code information, the trolley enters a simple homing program. The trolley adjusts its position to approach the X-axis and Y-axis and adjusts its angle so that there is only a lateral deviation S. The lateral deviation S is the perpendicular distance between the center points of the two wheels of the AGV and the ideal travel trajectory.

[0030] Step 2: Set the first reference deviation value M1, the second reference deviation value M2, and the third reference deviation value M3 according to the actual scenario, and M1>M2>M3. Select different deviation correction methods according to the magnitude of the lateral deviation.

[0031] When the lateral deviation S>M1, first select the rotation homing method for deviation correction. When the lateral error is corrected to be less than the second reference deviation value M2, if the lateral error is less than the third reference deviation value M3, the deviation correction is completed. If the lateral error is not less than the third reference deviation value M3, execute the deviation correction method once until the lateral error is less than the third reference deviation value M3.

[0032] When M1>S>M2, if the number of QR code nodes passed during deviation correction is greater than or equal to 2, the system selects the continuous deviation correction method. If the number of nodes passed during deviation correction is only 1, the system selects the low-speed single deviation correction method until the lateral error is less than the third reference deviation value M3.

[0033] When S<M2, the system selects the single deviation correction method until the lateral error is less than the third reference deviation value M3.

[0034] The rotation homing method includes:

[0035] Judge the current head direction. When the head points to the X-axis direction, the vehicle body rotates 90° in place. At this time, the head points to the Y-axis direction. Select forward or backward according to the magnitude of the Y coordinate to approach the X-axis. Stop after adjusting to a distance of ±Pmm from the X-axis. Then rotate to the initial X-axis direction and select forward or backward according to the magnitude of the X coordinate to approach the Y-axis. Stop after adjusting to a distance of ±Pmm from the Y-axis until the center of the vehicle body is within the circle with a radius of the third reference deviation value M3 from the center of the QR code and there is no angular deviation, and no deviation correction is required for straight travel. Here, P is the allowable deviation of the trolley from the coordinate axis.

[0036] When the head points to the Y-axis direction, adopt the same method.

[0037] The single deviation correction method includes:

[0038] After reading the current lateral deviation e and angular deviation θ, the data is transmitted to the fuzzy PID controller, which outputs the control parameter ΔV according to the input value and the expected value, thereby realizing the differential motion error elimination. During driving, the current lateral deviation e and angular deviation θ are calculated based on the digital micro-inertial navigation system MINS and the encoder to determine whether the correction is completed. If not, the value is returned to the input end, and the correction is cyclically performed until the correction is completed;

[0039] The continuous deviation correction method comprises:

[0040] During the first deviation correction, the maximum lateral deviation is limited to the second reference deviation value M2, and the angular deviation is negligible due to the effect of simple homing, and the deviation correction is performed using a one-time deviation correction method;

[0041] When the car reaches the second QR code, it reads the lateral deviation and angular deviation again, and directly calls the correction method once to complete the final correction amount.

[0042] Optionally, the simple homing procedure includes: after reading the current QR code value, determining whether the current vehicle head is pointing to the X-axis or the Y-axis;

[0043] When the front of the vehicle points to the X-axis direction, the vehicle will move straight at a low speed to approach the Y-axis and stop when it is ±Nmm away from the Y-axis. If the front of the vehicle points to the Y-axis direction, it will approach the X-axis and stop when it is ±Nmm away from the X-axis, where N is the distance error allowed by the homing program.

[0044] Determine the nearest angle that the current vehicle head is pointing to, including: 0°, 90°, 180°, 270°. Stop when the vehicle head differential fine-tuning points to the closest angle ±S°. If it is not in place at one time, adjust it repeatedly until the target angle is reached, where S is the angle error allowed by the homing program.

[0045] Optionally, the low-speed primary deviation correction method is to reduce the running speed of the AGV and then execute the primary deviation correction method.

[0046] The third object of the present invention is to provide an AGV control method based on two-dimensional code positioning and navigation, which is implemented based on the above-mentioned AGV control system based on two-dimensional code positioning and navigation, and adopts the above-mentioned AGV deviation correction method to realize the deviation correction of the vehicle movement process, including:

[0047] Step 1: Layout the QR code posting according to the usage scenario;

[0048] Step 2: After the actual scene layout of the QR code is completed, an electronic map is generated according to different code numbers in the host computer program and displayed on the host computer interface;

[0049] Step 3: The car is powered on, the AGV onboard system reads the valid QR code, and enters the simple homing procedure. After homing, the AGV onboard system obtains the current position information of the car and uploads the information to the host computer. The host computer interface updates the car information;

[0050] Step 4: The host computer obtains the current position information of the car and sends control instructions;

[0051] If the control instruction is a point-to-point instruction, the host computer performs optimal path iteration based on the current point and the target point through the ant colony method to generate a path data packet; if the control instruction is an action instruction, an action instruction packet is generated;

[0052] Step 5: The AGV vehicle-mounted system receives the path data packet and / or action instruction packet, executes the instruction, and the vehicle starts running;

[0053] Step 6: During the operation of the vehicle, the digital micro-inertial navigation system uses the above-mentioned AGV correction method to correct the lateral and angular deviations that occur during operation.

[0054] The beneficial effects of the present invention are:

[0055] The present invention proposes a new homing method, which solves the shortcomings of traditional AGVs that are in a silent state after power-on and cannot automatically and simply return to their original position, and achieves the purpose of automatically eliminating angle deviations at the start and simplifying the complexity of subsequent corrections; through a multi-strategy correction scheme, it solves the problems of the singleness and narrow application of traditional correction methods, and achieves the purpose of simplifying the correction method and effectively reducing the correction time; through multi-sensor data fusion, it solves the problems of single data acquisition and one-sided data acquisition in traditional correction schemes, which leads to the failure of correction strategies, and achieves the purpose of high-stability and high-precision operation correction.

[0056] The present invention optimizes the AGV initialization deviation to ensure that the AGV is always corrected with a small deviation posture. Secondly, the correction process is simplified through a multi-strategy correction method. Finally, the digital micro-inertial navigation system MINS and encoder information are integrated for track calculation to ensure the accuracy of the fuzzy PID return parameters, effectively optimize the AGV correction capability, effectively reduce the steps and time required for correction, and improve the stability of the AGV control system. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0058] Figure 1This is a general diagram of the AGV control system based on QR code positioning and navigation provided by the present invention.

[0059] Figure 2 This is a diagram of the host computer interface of the AGV control system based on QR code positioning and navigation provided by the present invention.

[0060] Figure 3 This is the overall drive control diagram of the AGV control system based on QR code positioning and navigation provided by the present invention.

[0061] Figure 4 This is a task allocation flow chart of the AGV control system based on QR code positioning and navigation provided by the present invention.

[0062] Figure 5 This is a flow chart of the homing and deviation correction of the AGV control system based on QR code positioning and navigation provided by the present invention.

[0063] Figure 6 A simple homing program diagram of the AGV control system based on QR code positioning and navigation provided by the present invention.

[0064] Figure 7 A diagram of the rotational homing method of the AGV control system based on QR code positioning and navigation provided by the present invention.

[0065] Figure 8 A diagram of a one-time deviation correction method for an AGV control system based on QR code positioning and navigation provided by the present invention.

[0066] Fig. 9 A diagram of a continuous deviation correction method for an AGV control system based on QR code positioning and navigation provided by the present invention.

[0067] Fig.10 This is a fuzzy PID control diagram of the AGV control system based on QR code positioning and navigation provided by the present invention.

[0068] Fig.11 This is a posture model diagram of the AGV control system based on QR code positioning and navigation provided by the present invention. DETAILED DESCRIPTION

[0069] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0070] Embodiment 1:

[0071] This embodiment provides an AGV control system based on two-dimensional code positioning and navigation, including: a host computer and an AGV vehicle-mounted system, wherein the host computer includes:

[0072] The map monitoring module is used to generate an electronic map based on the QR code label and update the vehicle's position and operating status in real time on the map;

[0073] Status information module, used to observe the status information and network connection information of the car in real time;

[0074] The command sending module is used to send action commands to the car;

[0075] The path planning module is used to search for the optimal path based on the starting point and target point of the car through the ant colony method, and finally obtain the optimal path;

[0076] The AGV vehicle-mounted system is used to realize the actual operation of the AGV; the AGV vehicle-mounted system includes: a main board and a slave board;

[0077] The main board includes: a real-time scheduling system for cyclically querying feedback information of different functional modules of the car. After receiving the feedback information, the main board packages the feedback information and uploads it to the host computer through the wireless communication module;

[0078] The slave board includes: a code reader, a digital micro-inertial navigation system and a motor drive module; after the code reader reads the current posture information, it is uploaded to the host computer, and the host computer sends a path data packet or a command data packet according to the current posture; after the slave board receives the path data packet or the command data packet, it starts to execute the command, and the motor drive module drives the motor to synchronously go straight to a specified distance or turn to a specified direction; the digital micro-inertial navigation system calculates the heading angle of the car based on the gyroscope and the accelerometer, and when the car passes through the blank area between two QR codes, it combines the feedback information of the motor encoder to infer the current posture of the car, and corrects the lateral and angular deviations that occur during operation. When the car passes through different ground QR codes, it continuously uploads posture information and continuously adjusts the posture to prevent error accumulation and ultimately cause code loss.

[0079] Embodiment 2:

[0080] This embodiment provides an AGV deviation correction method based on two-dimensional code positioning and navigation, which is implemented based on an AGV control system based on two-dimensional code positioning and navigation recorded in embodiment 1, and includes:

[0081] Step 1: After the AGV is powered on, the barcode reader reads the valid QR code information, and the trolley enters the simple homing procedure. The trolley adjusts its position to approach the X-axis and Y-axis, and adjusts the angle so that the trolley only has a lateral deviation S;

[0082] Step 2: according to the actual scene, a first reference deviation value M1, a second reference deviation value M2 and a third reference deviation value M3 are set, and M1>M2>M3, and different correction methods are selected according to the size of the lateral deviation;

[0083] When the lateral deviation S > M1, first select the rotation and homing method for deviation correction. When the lateral error is corrected to be less than the second reference deviation value M2, if the lateral error is less than the third reference deviation value M3, the deviation correction is completed. If the lateral error is not less than the third reference deviation value M3, execute the deviation correction method once until the lateral error is less than the third reference deviation value M3;

[0084] When M1 > S > M2, if the number of nodes passed during deviation correction is greater than or equal to 2, the system selects the continuous deviation correction method. If the number of nodes passed during deviation correction is only 1, the system selects the low-speed single deviation correction method until the lateral error is less than the third reference deviation value M3;

[0085] When S < M2, the system selects the single deviation correction method until the lateral error is less than the third reference deviation value M3;

[0086] The rotation and homing method includes:

[0087] Judge the current head pointing direction. When the head points to the X-axis direction, the vehicle body rotates 90° in place. At this time, the head points to the Y-axis direction. Select forward or backward according to the size of the Y coordinate to approach the X-axis. Stop after adjusting to ±P mm from the X-axis. Then rotate to point to the initial X-axis direction. Select forward or backward according to the size of the X coordinate to approach the Y-axis. Stop after adjusting to ±P mm from the Y-axis until the center of the vehicle body is within the circle with a radius of the third reference deviation value M3 from the center of the QR code and there is no angular deviation. Go straight without deviation correction. Here, P is the allowable deviation of the vehicle from the coordinate axis;

[0088] When the head points to the Y-axis direction, adopt the same method;

[0089] The single deviation correction method includes:

[0090] After reading the current lateral deviation e and angular deviation θ, input this data into the fuzzy PID controller. The fuzzy PID controller outputs the control parameter ΔV according to the input value and the expected value, so as to realize differential motion to eliminate the error. During driving, judge the current lateral deviation e and angular deviation θ based on MINS and the encoder to determine whether the deviation correction is completed. If not, return this value to the input end and perform cyclic deviation correction until the deviation correction is completed;

[0091] The continuous deviation correction method includes:

[0092] During the first deviation correction, limit the maximum lateral deviation to the second reference deviation value M2. The angular deviation is ignored due to the effect of simple homing. Adopt the single deviation correction method for deviation correction;

[0093] When the vehicle reaches the second QR code, read the lateral deviation and angular deviation again, and directly call the single deviation correction method to complete the final deviation correction amount.

[0094] Embodiment three:

[0095] This embodiment provides an AGV control system and a control method based on QR code positioning navigation. The system and method are described as follows:

[0096] like Figure 1 As shown, the control system of the present invention is divided into two parts: a host computer and an AGV vehicle-mounted system.

[0097] The host computer is built based on the JavaFX architecture, and its interface includes map monitoring, status information, command sending and path planning sections.

[0098] The AGV vehicle system (lower computer) is divided into a master and slave board. The two STM32 boards communicate through the serial port. The communication content mainly includes command information and action completion return information. The main board is mainly equipped with modules such as obstacle avoidance radar, wireless communication, power monitoring, command transmission and reception, and LCD display. The slave board is mainly equipped with modules such as code reader, micro inertial navigation system, and servo motor.

[0099] The overall workflow of the system is as follows: when the host computer sends a command for the car to reach the specified position, it searches for the optimal path through the ant colony algorithm based on the position information of the car after it goes online, and finally obtains the optimal path. The path package is composed of multiple nodes, each of which contains the address code ID, action code, speed gear and action parameters. This constitutes a series of complete action execution commands to ensure that the car reaches the specified position quickly and accurately.

[0100] The following is a detailed description of how the system works.

[0101] The QR codes are laid out according to the usage scenarios. The spacing and parallelism between the codes are required to be highly guaranteed, otherwise human errors will inevitably occur, resulting in poor running accuracy and stability of the trolley. In this embodiment, the QR code adopts a 4x4 array with a spacing of 1070mm between the two codes. This type of DM code can still be effectively used even if it is severely damaged, meeting a variety of application scenarios.

[0102] After the actual scene layout of the QR code is completed, an electronic map is generated according to different code numbers in the host computer program, and then displayed in the host computer interface. The map also needs to update the current location of the car and the charging pile information in real time, such as Figure 2 shown.

[0103] Figure 2 On the left is an electronic map constructed in a simple scenario, showing the QR code numbers and the charging location of the car using the charging pile standard. Figure 2The car icon in the lower left corner will automatically jump to the current car number position. Figure 2 The AGV status information in the middle and upper part will be updated, showing the current vehicle ID, operating status, power and voltage information, the current QR code number, and the deviation of the vehicle from the X and Y axes and the current pointing angle when it is above this code. This information will be used as the initial data source for correction later.

[0104] The two columns in the upper right corner are the server configuration and communication status information columns. The communication status information module displays the network connection status of the target car, and the server configuration module is the PC client network connection status column. The two are the PC login information and the AGV car login information respectively.

[0105] Figure 2 The lower right corner is the command sending panel, which operates as follows Figure 4 As shown in the figure, after obtaining the position information, the host computer can send control instructions, which are mainly divided into two command modes, one is point-to-point motion instructions, and the other is car motion instructions.

[0106] The first type of point-to-point motion is common. In the usage scenario, the task often performed is to complete the point-to-point motion. Figure 2 Point-to-point movement in the lower right corner, after filling in the designated arrival point, the host computer iterates the optimal path based on the current point and the target point through the ant colony method to generate a path data packet. The path data packet consists of multiple nodes, and each node contains four items. They are the address code ID (used to confirm whether the current code position is a command execution code), action code (forward, turning, lifting and rotation commands), speed gear (select different running speeds according to different actions) and action parameters (straight distance, turning angle, lifting height and rotation angle requirements). The host computer then adds timestamps and checksums according to the agreed command format. The encapsulated path data packet can be sent to the AGV vehicle system mainboard through the wireless communication module.

[0107] The AGV vehicle system mainboard is equipped with an obstacle avoidance radar module, a wireless communication module, a power communication module, a command transceiver module, and an LCD display module. The mainboard is equipped with a FreeRTOS system to schedule each task module in real time, and to query different module information in a round-robin manner, including reading the power module power and voltage information, the obstacle avoidance module object approach information, and the path information uploaded from the board and the command completion feedback. After receiving the corresponding information, the mainboard packages it and uploads it to the PC host computer through the wireless communication module. After receiving the information, the host computer continuously updates the interface information to visualize the AGV operation status. At the same time, it satisfies the host computer's scheduling and task allocation for the global AGV.

[0108] After power-on, the FreeRTOS system asks each task whether to enter the running state according to the order of task priority, while other tasks are in the suspended state. The obstacle avoidance radar module is set to generate an alarm only when it is 30cm close to the target, and forces the current AGV to stop running, effectively improving safety. The power module returns the current power information to prompt whether it needs to be charged to avoid power outage and crash. The LCD display module displays the power information, which can be viewed at close range and can also be viewed on the host computer. When the host computer sends a command or the AGV vehicle system sends a response return packet, the wireless communication module ensures stable and effective communication between the two. Since the AGV vehicle system is controlled by a master and slave board, the master and slave boards realize the command sending and receiving functions through serial communication, and its content includes path data packets and code reading information.

[0109] The AGV vehicle-mounted system is equipped with a code reader module, a digital micro-inertial navigation system module, and a drive module. The code reader reads the current position information according to the QR code posted on the ground, including the tag number, X-axis distance, Y-axis distance, and angle information. By reading the QR code information, the lower computer will upload a series of information to the upper computer. The upper computer interface updates the car information, showing its real-time position and current posture status, which is convenient for the upper computer to schedule and assign tasks for the AGV. The digital inertial navigation system can calculate the heading angle based on the gyroscope and accelerometer. When there is a blank area between the two QR codes, it combines the feedback information of the motor encoder to calculate the current AGV posture status. At the same time, it corrects the lateral and angular deviations that occur during operation to prevent the accumulation of errors from causing the failure of code reading, resulting in the loss of the path and safety accidents. The drive module adopts a differential drive method, which has the advantages of high flexibility and small turning radius. It is developed based on the CANopen protocol and has the advantages of error detection, support for multiple nodes, high real-time performance and low cost. The AGV car uses STM32 as a processor to manage multiple motors at the same time, such as Figure 3 As shown. Four slave stations are mounted under the master station, which are the left and right motors, the lifting motor and the rotating motor. CANopen can realize real-time control of multiple slave stations at the same time. The control process is as follows Figure 2 shown.

[0110] AGV correction steps are as follows Figure 5 After the AGV is powered on, if the read QR code is not empty, it means that a valid QR code is read, and the AGV enters the simple homing program. The program implementation steps are as follows: Figure 6As shown in the figure, after reading the current QR code value, determine whether the current head of the vehicle is pointing to the X-axis or the Y-axis. When the head of the vehicle is pointing to the X-axis, the car will move straight at a low speed close to the Y-axis and stop at ±3mm. If the head of the vehicle is pointing to the Y-axis, this action is close to the X-axis. After completing this action, determine which of the current head of the vehicle is closest to the angle of 0°, 90°, 180°, and 270°. The head of the vehicle will stop when the differential fine-tuning is closest to the angle of ±0.5°. If it is not in place at one time, it will be adjusted cyclically until the target is reached, and then the system will jump into the main loop function.

[0111] After receiving the path data packet or command packet sent by the host computer, the car makes the following choices based on the size of the lateral error. When the deviation is greater than 20mm, the AGV selects the rotation homing method. The implementation steps of this method are as follows: Figure 7 As shown, based on simple homing, the current vehicle head direction is determined. When the vehicle head points to the X-axis, the vehicle body rotates 90° on the spot. At this time, the vehicle head points to the Y-axis. According to the size of the Y-axis, it moves forward or backward, and stops after adjusting to ±2mm. Then it rotates to point to the initial direction. At this time, it points to the X-axis. According to the size of the X-axis, it moves forward or backward, and stops after adjusting to ±2mm. At this time, the center of the vehicle body is within a 2mm radius circle from the center of the QR code, and there is no angle deviation. No correction is required for straight driving. This type of elimination method has high requirements for steering accuracy and accuracy around the origin. It is often impossible to eliminate the lateral deviation to within 2mm at one time. Therefore, if it is greater than 10mm after elimination, repeat the above steps. Otherwise, it is determined whether it is less than 2mm and the correction is completed.

[0112] If it is less than 10mm but greater than 2mm, a single correction method is used to complete the final correction purpose. The implementation process of this method is as follows: Figure 8 As shown in the figure, after reading the current lateral deviation e and angular deviation θ, the data is passed to the fuzzy PID. The fuzzy PID outputs the control parameter ΔV according to the input value and the expected value, thereby realizing the differential motion to eliminate the error. During driving, the current e and θ are calculated based on the digital micro-inertial navigation system MINS and the encoder to determine whether the correction is completed. If not, the value is returned to the input end to continue a new cycle of correction until the correction is completed.

[0113] The fuzzy PID controller is constructed as follows: Fig.10 As shown, in this embodiment, the attitude deviation (including lateral deviation and angular deviation) is taken as input, and is passed into the fuzzy controller to optimize the appropriate kp, ki, and kd values, which are then used in the PID controller. The PID controller outputs an appropriate ΔV value according to the input deviation and the deviation change rate. Such ΔV value continuously adjusts the left and right wheel speeds according to its positive and negative values, and continuously corrects the lateral deviation and angular deviation. At the same time, the current error is calculated through the digital micro-inertial navigation system MINS (which is based on the integration of gyroscopes and accelerometers and can solve the attitude angle through the built-in algorithm. The heading angle is used in this embodiment) and the encoder. The calculation process is as follows:

[0114] Building an accurate kinematic model is the basis for designing navigation control algorithms. Fig.11 The schematic diagram of the AGV kinematic model is shown in Figure 2. Assume that the AGV performs differential motion at time t, and the right wheel speed v r Less than the left wheel speed v l , the trajectories of the left and right wheels are both circular arcs, the AGV turns right, and the velocity of the center of mass is v(v r 、v l The units of , v are both mm / s). Within the sampling time ΔT(s), the rotation angle is Δθ(rad):

[0115]

[0116] Where: D is the distance between the two wheels (mm); R is the radius of the driving path trajectory (mm),

[0117] R=(D(v r +v l )) / (2(v r -v l )).

[0118] AGV's velocity component v in the X and Y directions x (mm / s), v y The calculation formula of yaw angular velocity (mm / s) and yaw angular velocity ω (rad / s) is shown in formula (2):

[0119]

[0120] Specify the initial pose P = [x, y, θ] T =[x0,y0,θ0] T , we can get the relationship between the pose matrix and the left and right wheel speeds at any time (3):

[0121]

[0122] The above is the dead reckoning implementation formula, where the left and right wheel speeds v r With v l The motor built-in encoder can be used to read in real time. As mentioned above, MINS can return the heading angle, which can be used to return θ in real time. Then equation (3) can be realized.

[0123] The returned lateral deviation and angular deviation are compared with the set value (both lateral deviation and angular deviation are expected to be corrected to 0 as the set value), and are continuously passed into the fuzzy PID as new inputs, such as Fig.10 As shown, through continuous correction, it eventually fluctuates within a smaller range, and the correction algorithm is completed.

[0124] When the deviation is less than 20mm, determine whether it is greater than 10mm. If it is less than 10mm, use a correction method such as Figure 8 If it is greater than, consider whether the number of nodes passed by this correction is greater than or equal to 2. In the case of multiple nodes, the system selects the continuous correction method, which is as follows Fig. 9 As shown in the figure, because the lateral error is greater than 10mm and less than 20mm, and the complete path has more than two nodes, in order to improve the correction accuracy, a two-step strategy is adopted. During the first correction, the maximum lateral deviation is limited to 10mm. The angular deviation can be ignored due to the simple homing effect. It is passed into the correction method for correction. When the second QR code is reached, the lateral deviation and angular deviation are read again. At this time, the lateral deviation is about 10mm. At this time, the correction method is directly called to complete the final correction amount. This method can ensure that the deviation is corrected to a smaller range when the target point is reached. However, when there is only one node, the low-speed correction method is selected under the condition of comprehensive consideration of the error size and time. The low-speed correction method is to reduce the running speed of the AGV to ensure that the correction method completes a larger correction within a shorter distance. It is only used in this type of situation and can also meet the use needs. At the same time, the complexity of correction is reduced. At this point, the correction is completed.

[0125] The above is the implementation process and deviation correction process of the entire system, which to a certain extent makes up for the defects of the current AGV homing and deviation correction methods on the market.

[0126] This embodiment proposes a new homing algorithm. After the AGV is powered on, the AGV recognizes the QR code and automatically executes a simple homing procedure, automatically approaches the current pointing direction X-axis or Y-axis, and automatically adjusts the angle to point to the angle closest to the current direction. Different strategies are determined according to the size of the lateral deviation to perform error elimination actions, namely, one-time correction elimination, continuous correction elimination, and in-situ rotation elimination. Different strategies are made according to different situations to minimize the error elimination time.

[0127] This embodiment proposes a method of combining a digital micro-inertial navigation system and an encoder to realize the functions of deviation correction and path calculation. Since the AGV runs at a high speed, dual sensors are used to improve the stability of deviation correction. At the same time, given the limited reading rate of the code reader, it is easy to stop in time and run over when passing the stop or turn code at full speed. The encoder is used to calculate the distance to the next QR code, and the second speed is switched in advance when it is necessary to stop or turn, to ensure the accuracy of the stop.

[0128] Some steps in the embodiments of the present invention may be implemented using software, and the corresponding software program may be stored in a readable storage medium, such as a CD or a hard disk.

[0129] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An AGV deviation correction method based on QR code positioning and navigation, characterized in that: The AGV deviation correction method is implemented based on an AGV control system based on QR code positioning and navigation, and includes: Step 1: After the AGV is powered on, the code reader reads the valid QR code information, and the trolley enters a simple homing procedure. The trolley adjusts its position to approach the X-axis and Y-axis, and adjusts the angle so that the trolley only has a lateral deviation S, which is the vertical distance between the center point of the two wheels of the AGV and the ideal travel trajectory; Step 2: according to the actual scene, a first reference deviation value M1, a second reference deviation value M2 and a third reference deviation value M3 are set, and M1> M2> M3, and different correction methods are selected according to the size of the lateral deviation; When the lateral deviation S>M1, the rotation homing method is first selected for correction. When the lateral error is corrected to be less than the second reference deviation value M2, if the lateral error is less than the third reference deviation value M3, the correction is completed. If the lateral error is not less than the third reference deviation value M3, the correction method is executed once until the lateral error is less than the third reference deviation value M3. When M1 > S > M2, if the number of QR code nodes passed by the correction is greater than or equal to 2, the system selects the continuous correction method; if the number of nodes passed by the correction is only 1, the system selects the low-speed one-time correction method until the lateral error is less than the third reference deviation value M3; When S < M2, the system selects a correction method until the lateral error is less than the third reference deviation value M3; The rotation homing method comprises: Determine the current direction of the vehicle head. When the vehicle head points to the X-axis direction, the vehicle body rotates 90° on the spot. At this time, the vehicle head points to the Y-axis direction. According to the size of the Y coordinate, choose to move forward or backward to get close to the X-axis. Adjust to a distance of ±Pmm from the X-axis and stop. Then rotate to point to the initial X-axis direction. According to the size of the X coordinate, choose to move forward or backward to get close to the Y-axis. Adjust to a distance of ±Pmm from the Y-axis and stop until the center of the vehicle body is within the radius circle of the third reference deviation value M3 from the center of the QR code. There is no angle deviation, and no correction is required for straight driving. P is the allowable deviation of the vehicle from the coordinate axis. When the front of the vehicle points to the Y-axis direction, take the same approach; The one-time deviation correction method comprises: After reading the current lateral deviation e and angular deviation θ, the data is transmitted to the fuzzy PID controller, which outputs the control parameter ΔV according to the input value and the expected value, thereby realizing the differential motion error elimination. During driving, the current lateral deviation e and angular deviation θ are calculated based on the digital micro-inertial navigation system MINS and the encoder to determine whether the correction is completed. If not, the value is returned to the input end, and the correction is cyclically performed until the correction is completed; The continuous deviation correction method comprises: During the first deviation correction, the maximum lateral deviation is limited to the second reference deviation value M2, and the angular deviation is negligible due to the effect of simple homing, and the deviation correction is performed using a one-time deviation correction method; When the car reaches the second QR code, it reads the lateral deviation and angle deviation again, and directly calls the correction method once to complete the final correction amount; The AGV control system based on two-dimensional code positioning and navigation includes: a host computer and an AGV vehicle-mounted system, wherein the host computer includes: The map monitoring module is used to generate an electronic map based on the QR code label and update the vehicle's position and operating status in real time on the map; Status information module, used to observe the status information and network connection information of the car in real time; The command sending module is used to send action commands to the car; The path planning module is used to search for the optimal path based on the starting point and target point of the car through the ant colony method, and finally obtain the optimal path; The AGV vehicle-mounted system is used to realize the actual operation of the AGV; the AGV vehicle-mounted system includes: a main board and a slave board; The main board includes: a real-time scheduling system for cyclically querying feedback information of different functional modules of the car. After receiving the feedback information, the main board packages the feedback information and uploads it to the host computer through the wireless communication module; The slave board includes: a code reader, a digital micro-inertial navigation system and a motor drive module; after the code reader reads the current posture information, it is uploaded to the host computer, and the host computer sends a path data packet or a command data packet according to the current posture; after the slave board receives the path data packet or the command data packet, it starts to execute the command, and the motor drive module drives the motor to synchronously go straight to a specified distance or turn to a specified direction; the digital micro-inertial navigation system calculates the heading angle of the car based on the gyroscope and the accelerometer, and when the car passes through the blank area between two QR codes, it combines the feedback information of the motor encoder to infer the current posture of the car, and corrects the lateral and angular deviations that occur during operation. When the car passes through different ground QR codes, it continuously uploads posture information and continuously adjusts the posture to prevent error accumulation and ultimately cause code loss.

2. The AGV deviation correction method according to claim 1, characterized in that: The modules carried by the mainboard include: obstacle avoidance radar module, wireless communication module, power module, command transceiver module, LCD display module and FreeRTOS system; The real-time scheduling system adopts the FreeRTOS system; The feedback information includes: the power supply voltage information fed back by the power module, the obstacle approach information fed back by the obstacle avoidance radar module, the path information fed back by the slave board, and the command completion feedback information; The LCD display module displays the battery information of the vehicle; The command transceiver module realizes the command transceiver function between the main board and the slave board through serial port communication.

3. The AGV deviation correction method according to claim 1, characterized in that: The host computer sends the optimal path generation path data packet to the AGV vehicle-mounted system, and the path data includes a plurality of node information, and each node information includes: an address code ID, an action code, a speed gear information and an action parameter.

4. The AGV deviation correction method according to claim 1, characterized in that: The AGV vehicle-mounted system is built based on STM32, and the main board and slave board are STM32 boards.

5. The AGV deviation correction method according to claim 1, characterized in that: The motor drive module adopts a differential drive mode and is developed based on the CANopen protocol.

6. The AGV deviation correction method according to claim 1, characterized in that: The host computer is built based on the JavaFX architecture.

7. The AGV deviation correction method according to claim 1, characterized in that: The simple homing procedure includes: after reading the current QR code value, determining whether the current vehicle head is pointing to the X-axis or the Y-axis; When the front of the vehicle points to the X-axis direction, the vehicle will move straight at a low speed to approach the Y-axis and stop when it is ±Nmm away from the Y-axis. If the front of the vehicle points to the Y-axis direction, it will approach the X-axis and stop when it is ±Nmm away from the X-axis, where N is the distance error allowed by the homing program. Determine the nearest angle that the current vehicle head is pointing to, including: 0°, 90°, 180°, 270°. Stop when the vehicle head differential fine-tuning points to the closest angle ±S°. If it is not in place at one time, adjust it repeatedly until the target angle is reached, where S is the angle error allowed by the homing program.

8. The AGV deviation correction method according to claim 1, characterized in that: The low-speed primary deviation correction method is to reduce the running speed of the AGV and then execute the primary deviation correction method.

9. An AGV control method based on QR code positioning and navigation, characterized in that: The method adopts the AGV deviation correction method according to any one of claims 1 to 8 to realize deviation correction of the vehicle movement process, including: Step 1: Layout the QR code posting according to the usage scenario; Step 2: After the actual scene layout of the QR code is completed, an electronic map is generated according to different code numbers in the host computer program and displayed on the host computer interface; Step 3: The car is powered on, the AGV onboard system reads the valid QR code, and enters the simple homing procedure. After homing, the AGV onboard system obtains the current position information of the car and uploads the information to the host computer. The host computer interface updates the car information; Step 4: The host computer obtains the current position information of the car and sends control instructions; If the control instruction is a point-to-point instruction, the host computer performs optimal path iteration based on the current point and the target point through the ant colony method to generate a path data packet; if the control instruction is an action instruction, an action instruction packet is generated; Step 5: The AGV vehicle-mounted system receives the path data packet and / or action instruction packet, executes the instruction, and the vehicle starts running; Step 6: During the operation of the vehicle, the digital micro inertial navigation system corrects the lateral and angular deviations that occur during the operation by using the AGV correction method as described in any one of claims 1 to 8.

Images