Method and system for controlling consistency of multiple unmanned forklifts and medium

Abstract

The invention discloses a control method and system for consistency of multiple unmanned forklifts and a medium. The method comprises the steps that a dynamic calibration board is used for conducting external parameter calibration on laser radars arranged on the tops of all the unmanned forklifts; each calibrated unmanned forklift is controlled to scan a target area to obtain a reference point cloud, actual control center offset parameters are determined in combination with the vehicle body model, and the relative position relation between the laser radar and the actual control center is determined; acquiring target point information, and calculating a conversion point corresponding to a target point in combination with the relative position relationship and the vehicle body model; and controlling each unmanned forklift to navigate to the switching point. According to the technical scheme, multiple unmanned forklifts of different models can be controlled at the same time to achieve the consistency effect, the consistency problem of multiple unmanned forklifts of the same or different models in digital coordinates and physical positions is solved, and the actual requirements are better met.

Description

Technical Field

[0001] The present invention relates to the field of unmanned forklift control, and in particular to a control method, system and medium for consistency of multiple unmanned forklifts. Background Art

[0002] In recent years, the application of unmanned forklifts in the field of logistics and handling has been gradually popularized with its good cargo carrying capacity, effectively improving the efficiency and accuracy of logistics and handling and reducing the workload of manpower. In order to cope with the large-scale cargo volume in logistics and handling, logistics centers, warehouses, sorting centers and other scenes have adopted a large number of unmanned forklifts for collaborative operation, which greatly improved the efficiency of cargo circulation.

[0003] In the process of unmanned forklift operation, in order to ensure that it can complete the logistics handling task safely and accurately, it is necessary to use high-precision sensors such as laser radar to build real-time maps of the operation scene. By emitting laser beams and receiving the reflected signals, laser radar can accurately measure the distance, shape and position information of objects in the surrounding environment, thereby constructing a three-dimensional spatial model, that is, an environmental map.

[0004] When multiple unmanned forklifts are operating in the same scene, building a separate map for each unmanned forklift can ensure the operating control accuracy of each unmanned forklift. However, this method is not only time-consuming and costly, but also greatly limits the efficiency and flexibility of logistics operations.

[0005] Multiple unmanned forklifts share a built environment map, which can effectively reduce the cost of map construction, simplify the operation process and improve the efficiency of multi-vehicle dispatching. However, due to the differences in the mechanical installation structure of each unmanned forklift, when navigating to the same target point on the map, there will often be position deviations in the actual physical space.

[0006] Therefore, when multiple unmanned forklifts share a scene map in the same scenario, how to ensure the accurate navigation of each unmanned forklift and achieve consistency among multiple vehicles, both aligning them in digital coordinates and synchronizing them in physical space, has become a key issue that needs to be urgently solved in the application of unmanned forklifts in the field of intelligent logistics. Summary of the invention

[0007] The present invention provides a control method, system and medium for the consistency of multiple unmanned forklifts, which can control the consistency of multiple unmanned forklifts of different models, solve the problem of inconsistency in digital coordinates or physical positions of multiple unmanned forklifts of the same or different models, and better meet practical needs.

[0008] According to one aspect of the present invention, a method for controlling consistency of multiple unmanned forklifts is provided, comprising:

[0009] Use a dynamic calibration plate to calibrate the external parameters of the laser radar configured on the top of each unmanned forklift;

[0010] After control calibration, each unmanned forklift scans the target area to obtain the reference point cloud, determines the actual control center offset parameters based on the vehicle model, and determines the relative position relationship between the laser radar and the actual control center;

[0011] Acquire target point information, and calculate a conversion point corresponding to the target point in combination with the relative position relationship and the vehicle model;

[0012] Control each unmanned forklift to navigate to the transfer point.

[0013] Optionally, use a dynamic calibration board to calibrate the external parameters of the laser radar configured on the top of each unmanned forklift, including:

[0014] The dynamic calibration plate includes N reflection marking points (N is a positive integer greater than or equal to 2), and the reflection marking points are arranged according to a predetermined geometric pattern;

[0015] The dynamic calibration plate is scanned by a laser radar, and the external parameters of the laser radar are calculated based on the scanning result and the known position relationship of the reflection mark point.

[0016] Optionally, the dynamic calibration plate is scanned by the laser radar, and the external parameters of the laser radar are calculated based on the scanning results and the known position relationship of the reflection mark points, including:

[0017] Control the dynamic calibration plate to move within a predetermined range, and during the movement, continuously collect point cloud data of the dynamic calibration plate through the laser radar;

[0018] Extracting geometric features of the dynamic calibration plate from the laser radar point cloud data;

[0019] A nonlinear optimization algorithm is used to calculate the extrinsic parameters of the LiDAR based on the acquired point cloud data and the known geometric features of the dynamic calibration plate.

[0020] Optionally, each unmanned forklift after control calibration scans the target area to obtain a reference point cloud, determines the actual control center offset parameters in combination with the vehicle body model, and determines the relative position relationship between each unmanned forklift laser radar and the actual control center, including:

[0021] Extract feature points representing the contour of the fork arm of the unmanned forklift from the reference point cloud;

[0022] A straight line fitting is performed on the characteristic points of the fork arm profile, and the offset parameters of the actual control center of each unmanned forklift relative to the theoretical control center are calculated based on the fitting results and the vehicle body model.

[0023] Optionally, obtaining target point information and calculating a conversion point corresponding to the target point in combination with a relative position relationship and a vehicle body model includes:

[0024] According to the relative position relationship between each unmanned forklift laser radar and the actual control center, determine the reference position point of each unmanned forklift laser radar;

[0025] According to the reference position point of each unmanned forklift laser radar and the vehicle model, the conversion point corresponding to the target point is determined.

[0026] Optionally, according to the reference position point of each unmanned forklift laser radar and the vehicle model, the conversion point corresponding to the target point is determined, including:

[0027] According to the constraints of the laser radar in the vehicle model and the theoretical control center in the vehicle model, the reference position information of the theoretical control center in the vehicle model is determined when the laser radar is at the reference position point;

[0028] The conversion point corresponding to the target point is determined according to the reference position information of the theoretical control center.

[0029] Optionally, the conversion point corresponding to the target point is determined according to the reference position point of each unmanned forklift laser radar and the vehicle model, and also includes:

[0030] Based on the laser radar external calibration parameters, determine the reference position point of the laser radar in the vehicle coordinate system when the actual control center of the vehicle is at the target point;

[0031] Based on the vehicle model, the theoretical control center position of the vehicle body in the laser radar coordinate system is calculated; the theoretical control center position of the vehicle body in the laser radar coordinate system is converted to the global map coordinate system to obtain the conversion point corresponding to the target point.

[0032] Optionally, the consistency control method for multiple unmanned forklifts further includes establishing a mapping table of target points and conversion points;

[0033] Before obtaining the target point information and calculating the conversion point corresponding to the target point by combining the relative position relationship and the vehicle model, the following steps are also included:

[0034] When obtaining new target point information, the mapping table is queried first to obtain the corresponding conversion point;

[0035] If there is no target point match, the calculated conversion point corresponding to the target point is added to the mapping table.

[0036] According to another aspect of the present invention, a control system for consistency of multiple unmanned forklifts is provided, comprising:

[0037] A storage unit, used to store the program of the steps of the consistency control method for multiple unmanned forklifts described in any embodiment of the present invention, so that the data acquisition unit, the control unit, and the processing unit can call and execute it in a timely manner;

[0038] A control unit, used to control the dynamic calibration plate to move within a predetermined range when performing external parameter calibration on the laser radar configured on the top of each unmanned forklift using the dynamic calibration plate; and control each unmanned forklift to navigate to a corresponding conversion point of the target point;

[0039] A data acquisition unit is used to collect laser point cloud data of a dynamic calibration plate when performing external parameter calibration on a laser radar configured on the top of each unmanned forklift using the dynamic calibration plate; and to collect point cloud data of a target area scanned by the laser radar when determining an actual control center offset parameter of each unmanned forklift according to a reference point cloud of a target area scanned by each unmanned forklift and a vehicle body model;

[0040] The processing unit is used to calculate the laser radar extrinsic calibration parameters based on the dynamic calibration plate point cloud data collected by the laser radar and the known geometric features of the dynamic calibration plate; calculate the actual control center offset parameters of each unmanned forklift according to the reference point cloud of the target area scanned by each unmanned forklift and the vehicle model, and calculate the relative position relationship between each unmanned forklift laser radar and the actual control center according to the actual control center offset parameters; calculate the conversion point corresponding to the target point according to the relative position relationship, target point information and the vehicle model.

[0041] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for controlling consistency of multiple unmanned forklifts as described in any embodiment of the present invention when executed.

[0042] The technical solution of the embodiment of the present invention uses a dynamic calibration plate to perform external parameter calibration on the laser radar configured on the top of each unmanned forklift; each unmanned forklift after calibration is controlled to scan the target area to obtain the reference point cloud of the target area. According to the fitting result of the reference point cloud, the actual control center offset parameter of each unmanned forklift is determined in combination with the body model of each unmanned forklift, and the relative position relationship between each unmanned forklift laser radar and the actual control center is determined according to the offset parameter. The target point information is obtained, and the conversion point corresponding to the target point is calculated in combination with the relative position relationship and the body model; each unmanned forklift is controlled to navigate to the conversion point corresponding to the target point.

[0043] This technical solution uses a dynamic calibration plate to calibrate the laser radar external parameters of each unmanned forklift, which can obtain diverse geometric information in a single calibration process, improve the robustness and accuracy of the laser radar external parameter calibration, and is more in line with the operating environment of the unmanned forklift. At the same time, according to the difference between the actual control center of each unmanned forklift and the theoretical control center of the vehicle model among multiple unmanned forklifts, the navigation target point is converted into a conversion target point that adapts to its own differences, and the differences of each vehicle are compensated. For the same target point, the actual control centers of multiple unmanned forklifts can be controlled to reach the same physical position. This system can adapt to unmanned forklifts of different models and different usage states to achieve the effect of multi-vehicle control consistency, which better meets actual needs.

[0044] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. 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.

[0046] Figure 1 is a flow chart of a method for consistency control of multiple unmanned forklifts provided according to Embodiment 1 of the present invention;

[0047] Figure 2 is a flow chart of calibrating laser radar extrinsic parameters using a dynamic calibration plate according to Embodiment 1 of the present invention;

[0048] Figure 3 is a schematic diagram of a point cloud of a target area where a laser radar scanning fork arm is located according to Embodiment 1 of the present invention;

[0049] Figure 4 It is a flow chart showing an optimization and improvement of a method for controlling consistency of multiple unmanned forklifts provided in Embodiment 2 of the present invention based on Embodiment 1;

[0050] Figure 5 This is a block diagram of a consistency control system for multiple unmanned forklifts provided according to Embodiment 3 of the present invention. DETAILED DESCRIPTION

[0051] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0052] It should be noted that the terms "first", "second", "target", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0053] Embodiment 1

[0054] When unmanned forklifts are operating in indoor warehousing, factory automation, outdoor ports, logistics parks and other scenarios, the lidar installed on the top has a wide field of view and optimal perception range, which can provide reliable positioning and navigation support for unmanned forklifts. When multiple unmanned forklifts are operating in the same scenario, in order to simplify the deployment process and reduce the cost of map construction, multiple unmanned forklifts often share a built environmental positioning map.

[0055] It should be noted that the internal navigation and control of the vehicle body of the unmanned forklift are based on the vehicle body model of the unmanned forklift, which can be obtained according to the vehicle body design parameters. The operation of the unmanned forklift is controlled by calculating the error-free vehicle body model parameters. However, there are differences between the actual mechanical installation parameters and the vehicle body model. When the navigation of the unmanned forklift is controlled by calculating the error-free vehicle body model parameters, when each unmanned forklift navigates to the same target point, it is difficult to maintain consistency in the spatial physical position reached by each unmanned forklift. At the same time, there are differences between the installation parameters of the laser radar configured on the top of the unmanned forklift and the theoretical error-free parameters of the laser radar in the vehicle body model. Even the laser radar installation parameters of each unmanned forklift of the same model are different, which will cause differences in the digital coordinates output by each unmanned forklift at the same point on the map.

[0056] Figure 1This is a flow chart of a method for controlling the consistency of multiple unmanned forklifts provided in the first embodiment of the present invention. This embodiment calibrates the laser radar positioning digital coordinates of multiple unmanned forklifts at the same target point, and controls the consistency of the physical positions reached by multiple unmanned forklifts operating in the same environment when navigating to the same target point on the map. The method can be executed by the control device of the unmanned forklift, and the control device of the unmanned forklift can be implemented in the form of hardware and / or software. Figure 1 As shown, the method includes:

[0057] S10 uses a dynamic calibration plate to calibrate the external parameters of the laser radar installed on the top of each unmanned forklift.

[0058] Specifically, to control the consistency of multiple unmanned forklifts, it is necessary to ensure the consistency of the digital coordinates output by each unmanned forklift at the same target point on the map, which requires the calibration of the laser radar external parameters of each unmanned forklift. The dynamic calibration plate in the embodiment of this technical solution includes N reflective marking points (N is a positive integer greater than 2), and the reflective marking points are arranged according to a predetermined geometric pattern. The dynamic calibration plate can use highly reflective materials and can be well recognized by the laser radar.

[0059] The S20 uses a laser radar to scan a dynamic calibration plate and calculates the external parameters of the laser radar based on the scanning results and the known position relationship of the reflection mark points.

[0060] The use of a dynamic calibration plate can obtain a variety of geometric information in a single calibration process, improving the accuracy and robustness of the lidar external parameter calibration. In particular, the use of a dynamic calibration plate can simulate the operating environment of an unmanned forklift, such as Figure 2 As shown in the figure, the specific process of using the dynamic calibration board for external parameter calibration is as follows:

[0061] S201 controls the dynamic calibration plate to move within a predetermined range, and during the movement, continuously collects point cloud data of the dynamic calibration plate through a laser radar.

[0062] Specifically, during calibration, the unmanned forklift is parked in a flat area, the laser radar on the top of the unmanned forklift is kept turned on, and the laser radar continuously collects point cloud data of the dynamic calibration plate. The dynamic calibration plate is placed within the scanning range of the laser radar, and it is ensured that the laser radar can clearly scan all the reflection mark points of the dynamic calibration plate. The dynamic calibration plate is controlled to translate or rotate within the field of view of the laser radar, and the calibration plate is moved to obtain scanning data at multiple different angles and positions. During the scanning process, the laser radar of the unmanned forklift remains stable and stationary.

[0063] S202 synchronizes the dynamic calibration plate and the laser radar scanning timestamp, and aligns the laser radar data with the position information of the dynamic calibration plate.

[0064] The position of each reflective marker on the dynamic calibration plate in the calibration plate coordinate system is usually represented by its coordinates (x, y, z), where x, y, and z represent the offset of the reflective marker along three orthogonal axes (usually the horizontal X axis, vertical Y axis, and depth Z axis) in the calibration plate coordinate system. These coordinate values ​​are fixed and will not change due to the movement of the calibration plate or the scanning of the laser radar.

[0065] During the scanning process, the laser radar will record the precise timestamp of each data point, and the dynamic calibration board will also record its precise position information and posture information as well as the corresponding timestamp. According to the timestamp information recorded by the laser radar, the timestamp closest to the laser radar timestamp is determined from the timestamp recorded by the dynamic calibration board, and the data scanned by the laser radar is precisely aligned with the position information of the dynamic calibration board.

[0066] S203 extracts geometric features of the dynamic calibration plate from the laser radar point cloud data.

[0067] Use clustering algorithms (such as DBSCAN) or threshold segmentation methods based on reflection intensity to segment the point cloud data into different areas or objects to better identify the dynamic calibration plate. Segment the points of the dynamic calibration plate from the original point cloud data collected by the lidar.

[0068] The geometric features of the dynamic calibration plate are extracted from the segmented point cloud data, and the edges and corners of the dynamic calibration plate are identified using methods based on geometric properties (such as curvature, normal vector, etc.). For example, the edges or corners of the dynamic calibration plate are detected by calculating the curvature or normal vector change of each point in the point cloud. Based on the extracted geometric features, the least squares method, RANSAC algorithm, and other methods are used to fit the dynamic calibration plate model parameters to obtain the dynamic calibration plate point cloud.

[0069] S204 uses a nonlinear optimization algorithm to calculate the external parameters of the laser radar based on the collected point cloud data and the known geometric features of the dynamic calibration plate.

[0070] Specifically, the transformation relationship between the dynamic calibration plate coordinate system and the laser radar coordinate system is established. The coordinates of the reflection mark point in the dynamic calibration plate coordinate system are (x b ,y b , z b ); the coordinates of the corresponding point in the laser radar coordinate system are (x l ,y l , z l );

[0071] The relationship between them can be expressed by the following formula:

[0072] Where R is a 3x3 rotation matrix, T = [t x ,ty,tz ] is a 3x1 translation vector and O is a 1x3 zero vector.

[0073] Reflection mark point Pb = [x b y b z b 1];

[0074] The coordinates of the corresponding point in the laser radar coordinate system Pl = [x l y l z l 1]

[0075] Pl=T*Pb

[0076]

[0077] According to the initial position relationship between the unmanned forklift's lidar and the dynamic calibration plate, the lidar is estimated using initial extrinsic parameters. The next step is to fine-tune the extrinsic parameters using nonlinear optimization algorithms such as Levenberg-Marquardt.

[0078] The present application scheme uses a nonlinear optimization algorithm (such as the Levenberg-Marquardt algorithm) to solve the optimal external parameter matrix. The optimization objective function is:

[0079] in, and They represent the point in the i-th lidar coordinate system and the point in the calibration plate coordinate system respectively, R is the rotation matrix, T is the translation vector, and ||.|| represents the Euclidean distance.

[0080] The dynamic calibration plate is scanned by the laser radar, and the external parameters of the laser radar are calculated based on the scanning results and the known position relationship of the reflection mark points. Using the dynamic calibration plate to calibrate the external parameters of the laser radar of the unmanned forklift can obtain data from more perspectives in a single operation, provide more comprehensive calibration information, and be closer to the actual use scenario of the unmanned forklift.

[0081] The above method can be used to calibrate the external parameters of the laser radar of each unmanned forklift to ensure the consistency of the digital coordinates of the laser radar of each unmanned forklift at the same point in the map.

[0082] In fact, the consistency of controlling multiple unmanned forklifts not only includes the consistency of the digital coordinates of the positioning of multiple unmanned forklifts at the same target point, but also includes the consistency of controlling the arrival of multiple unmanned forklifts, specifically including:

[0083] S30 controls each calibrated unmanned forklift to scan the target area to obtain the reference point cloud.

[0084] The target area refers to the area where the fork arm of the unmanned forklift is located. The area is scanned by the laser radar to obtain point cloud data containing the contour information of the fork arm. The laser radar of each unmanned forklift after calibration is used to scan the area where the fork arm of the unmanned forklift is located to obtain the laser radar point cloud of the area where the fork arm is located.

[0085] S40 determines the actual control center offset parameters in combination with the vehicle body model, and determines the relative position relationship between the laser radar and the actual control center.

[0086] The laser radar is configured on the top of the unmanned forklift and can scan the point cloud of the target area including the fork arm of the unmanned forklift. The feature points of the reference point cloud of the target area representing the contour of the fork arm of the unmanned forklift are extracted, and a straight line is fitted to the feature points of the fork arm contour. The actual control center offset parameters of each unmanned forklift are determined based on the vehicle body model.

[0087] The vehicle body model is an ideal model built based on the design parameters of the unmanned forklift, including the size, wheelbase, and rear wheel spacing of the unmanned forklift. During the driving process, the unmanned forklift needs to be constantly positioned and navigated. Generally, the vehicle control center point is selected to help the unmanned forklift adjust its position and direction in real time to ensure accurate driving along the predetermined path.

[0088] By selecting the control center point of the unmanned forklift, a more flexible and adjustable path planning scheme can be constructed. When encountering obstacles, road changes or other emergencies, the unmanned forklift can adjust its driving strategy according to the position of the control center point to ensure the stable operation of the system. The center axis of the unmanned forklift's rear wheels can be selected as the vehicle's control center point. Controlling the unmanned forklift to navigate to the target point is essentially controlling the control center point of the unmanned forklift to stop at the target point.

[0089] Considering that the laser radar installed on the top of different models of unmanned forklifts has different external parameters, such as Figure 3 As shown in the figure, for the case where the laser radar can scan the rear wheels of the unmanned forklift, the rear wheel position of the unmanned forklift can be directly determined based on the scanned laser point cloud, and then the actual control center point of the unmanned forklift can be obtained by calculating the center axis of the two rear wheels; for the model where the top laser radar cannot scan the rear wheels of the unmanned forklift, the control center point can be determined by combining the scanned fork arm point cloud and the body model parameters. By comparing the fork arm contour in the reference point cloud with the theoretical control center position in the body model, the offset parameter of the actual control center relative to the theoretical control center is calculated.

[0090] More specifically, the fork arm is installed perpendicular to the vehicle body in the body model parameters, but due to the processing accuracy and installation accuracy of the fork arm, the actual installation of the fork arm may deviate from the ideal situation. The point cloud of the fork arm area is scanned by laser radar, and the point cloud of the fork arm area is clustered and segmented using the DBSCAN algorithm. The feature point cloud of the fork arm is extracted, and the extracted fork arm point cloud is linearly fitted to determine the actual installation deviation of the fork arm. The actual installation position of the fork arm is determined based on the body model parameters and the actual deviation value of the fork arm, and finally the offset parameter of the actual control center of the unmanned forklift relative to the theoretical control center of the body model is determined.

[0091] Among them, the method for obtaining the offset parameters of the actual control center of the unmanned forklift relative to the control center of the vehicle model includes: taking the theoretical control center of the vehicle model as the origin, that is, taking the center of the rear wheel in the vehicle model as the origin to establish a three-dimensional coordinate system. Among them: the positive direction of the X-axis is the direction of the front of the vehicle, that is, the forward direction of the unmanned forklift. The positive direction of the Y-axis is the left side of the vehicle body, perpendicular to the X-axis, and points to the left side of the vehicle. The positive direction of the Z-axis is perpendicular to the ground and extends upward to form a height coordinate in three-dimensional space. The laser radar is represented as (x1,0,z1) in this coordinate system, where x1 represents the offset of the laser radar on the X-axis (that is, the direction of the front of the vehicle) (a positive value represents a forward offset, and a negative value represents a backward offset). Since the laser radar is located on the center line of the vehicle, the Y-axis coordinate is 0, and z1 represents the position of the laser radar on the Z-axis (that is, the vertical height), indicating the height of the laser radar from the ground.

[0092] A straight line fitting algorithm is used to perform straight line fitting on the extracted fork arm feature point cloud, and the fitted straight line is converted to the vehicle model coordinate system to determine the actual installation position of the unmanned forklift fork arm and the actual control center of each unmanned forklift. The actual control center of each unmanned forklift and the position of the lidar are converted to the vehicle model coordinate system. The relative position relationship between each unmanned forklift lidar and the actual control center of the unmanned forklift is determined in the same vehicle model coordinate system based on the offset parameters of the actual control center of each unmanned forklift relative to the control center of the vehicle model and the lidar external parameter calibration information.

[0093] S50 obtains target point information, and calculates a conversion point corresponding to the target point in combination with the relative position relationship and the vehicle body model.

[0094] When each unmanned forklift is running in the scene, it obtains task information including the target ID of the task to be executed from the server, and obtains the target point information of the specific task after task analysis. If the difference in mechanical installation parameters is not considered, the vehicle model will navigate according to the target point information, and the vehicle model control center of the unmanned forklift is expected to stop at the target point; however, there is a deviation between the actual mechanical installation parameters of the unmanned forklift and the vehicle model parameters. In the embodiment of the present invention, the target point of each vehicle is converted according to its own differences.

[0095] According to the relative position relationship between each unmanned forklift laser radar and the actual control center, the reference position point of each unmanned forklift laser radar is determined. Specifically, controlling the unmanned forklift to navigate to the target point is essentially to expect the actual control center of each unmanned forklift to reach the target point. According to the relative position relationship between each unmanned forklift laser radar and the actual control center, it can be determined that when the actual control center of the unmanned forklift is at the target point, the reference position point P of each unmanned forklift laser radar is laser .

[0096] Further, the conversion point corresponding to the target point is determined based on the reference position point of each unmanned forklift laser radar and the vehicle model. Specifically, it includes: based on the laser radar external parameter calibration parameters, determine the reference position point of the laser radar in the vehicle coordinate system when the actual control center of the vehicle is at the target point; convert the reference position point of the laser radar in the vehicle coordinate system to the laser radar coordinate system through external parameters, and calculate the position point of the theoretical control center of the vehicle in the laser radar coordinate system; convert the theoretical control center position point of the vehicle model in the laser radar coordinate system to the global map coordinate system to obtain the conversion point corresponding to the target point.

[0097] The vehicle model stores the constraints between the laser radar and the model theoretical control center, which determines that the laser radar is at the reference position point P. laser When the reference position information P of the theoretical control center of the vehicle model is theo-con According to the vehicle model theory, the reference position information P of the control center theo-con Determine the conversion point corresponding to the target point. Based on the laser radar external parameter calibration parameters, determine the reference position point of the laser radar in the vehicle body coordinate system when the actual control center of the vehicle body is at the target point; based on the vehicle body model, calculate the reference position point of the theoretical control center of the vehicle body model in the laser radar coordinate system. Convert the theoretical control center position point of the vehicle body model in the laser radar coordinate system to the global map coordinate system to obtain the conversion point corresponding to the target point.

[0098] More specifically, after the unmanned forklift obtains the target ID from the server, it determines the target point through task analysis and unifies it in the laser radar coordinate system. It is expected that the actual control center of each vehicle will reach the target point P true-con According to the relative position relationship between the actual control center of the unmanned forklift and the laser radar, the position point P of the laser radar of the unmanned forklift is determined when the actual control center of the unmanned forklift navigates to the target point. laser The navigation of unmanned forklift is based on the body model parameters, that is, the error-free standard parameters, which are obtained according to the body design parameters of each unmanned forklift, including at least the length, width and height of the unmanned forklift, as well as the distance between the rear wheels and the distance between the axles.laser , combined with the vehicle model parameters, calculate the location point P of the theoretical control center of the unmanned forklift in the model theo-con , P in the laser radar coordinate system theo-con Convert to the global map coordinate system to get the conversion point P' corresponding to the target point theo-con It can be understood that when the unmanned forklift navigates to the conversion point P' on the map according to the vehicle model parameters theo-con At this time, in the laser radar coordinate system, the theoretical control center of the unmanned forklift model reaches P theo-con , then according to the body model parameters, it can be determined that when the theoretical control center of the body model of the unmanned forklift reaches the laser radar coordinate system P theo-con The laser radar of the unmanned forklift is located at position P laser Furthermore, according to the relative position relationship between the laser radar of the unmanned forklift and the actual control center, it can be determined that the actual control center of the unmanned forklift is located at position point P true-con In this way, the physical consistency of multiple unmanned forklifts when they run to the same target point on the map is ultimately ensured.

[0099] S60 controls each unmanned forklift to navigate to the conversion point corresponding to the target point.

[0100] The vehicle body model is passed according to the vehicle body design parameters of each unmanned forklift. The unmanned forklift calculates navigation and controls the operation of the unmanned forklift based on the theoretical control model. Through the laser radar position conversion method, the reference position point of the laser radar is first determined when the actual control center of each unmanned forklift is expected to reach the target point. Then, based on the rear wheel spacing in the vehicle body model and the laser radar external parameter information, the position of the theoretical control center point in the vehicle body model is determined. Then, the position of the theoretical control center point is converted to the map coordinate system, and the conversion point corresponding to the target point can be obtained. The unmanned forklift is controlled to navigate to the conversion point, which can ensure that the actual control center of each vehicle can eventually accurately reach the predetermined target point position.

[0101] The technical solution disclosed in the embodiment of the present invention uses a dynamic calibration plate to calibrate the external parameters of the laser radars of multiple unmanned forklifts to ensure the positioning consistency of the multiple unmanned forklifts, that is, the digital coordinates of the same target point in the map are consistent; the laser point cloud of the target area is obtained according to the laser radar, and the fork arm point cloud features are extracted from the laser point cloud of the target area, and the extracted fork arm point cloud is fitted into a straight line, which is compared with the body model parameters to determine the deviation between the actual fork arm installation and the fork arm parameters in the model, and thereby determine the difference between the actual control center of the unmanned forklift and the model control center, and determine the relative position relationship between the actual control center and the laser radar. By calculating the position point of the model control center when the actual control center is at the target point, the conversion point corresponding to the target point is obtained, and the unmanned forklift is controlled to navigate to the conversion point to ensure that the actual control centers of multiple unmanned forklifts reach the target point, and achieve consistency in the arrival of multiple unmanned forklifts.

[0102] The embodiment of the present invention obtains the laser point cloud of the fork arm of each unmanned forklift, determines the actual installation position of the fork arm of each unmanned forklift according to the body model of the unmanned forklift, and finally determines the actual control center position of each unmanned forklift. The laser radar configured on the top of the unmanned forklift is used to scan the fork arm area of ​​the unmanned forklift to obtain the laser point cloud of the fork arm area, clustering and segmentation are performed from the obtained laser point cloud, and the feature point cloud of the fork arm of the unmanned forklift is extracted. The feature point cloud of the fork arm is fitted into a straight line, and compared with the fork arm installation parameters in the body model, so as to determine the deviation between the actual fork arm installation parameters of each unmanned forklift and the fork arm installation parameters in the body model.

[0103] Generally, the rear wheel center of the unmanned forklift is selected as the control center of the unmanned forklift. The deviation between the actual fork arm installation parameters of each unmanned forklift and the theoretical fork arm installation parameters in the vehicle model is determined by the above method, and then the actual control center position point of each unmanned forklift is obtained. According to the actual control center and the laser radar external parameter calibration parameters, the relative position relationship between the actual control center of the unmanned forklift and the laser radar is determined.

[0104] The advantages of the method of the present invention are that the dynamic calibration plate is used for external parameter calibration, which improves the calibration accuracy and efficiency; by calculating the conversion point, the individual differences of each unmanned forklift are compensated, and the consistency control of multiple vehicles on the same map is achieved; it is suitable for unmanned forklifts of different models and different usage conditions, the operation process is simple, and it has wide practicality.

[0105] In practical applications, this method can significantly improve the collaborative efficiency and accuracy of multiple unmanned forklifts in logistics warehousing, factory automation and other scenarios. For example, in a large logistics center, multiple unmanned forklifts of different models can accurately hand over goods at the same pickup or unloading point, reducing manual intervention and the probability of error.

[0106] Embodiment 2

[0107] Figure 4 The flowchart of a method for controlling consistency of multiple unmanned forklifts provided in the second embodiment of the present invention is optimized based on the above embodiment. The optimization is to establish a mapping table of target points and conversion points. Figure 4 As shown, the method of this embodiment includes the following steps:

[0108] Before obtaining the target point information and calculating the conversion point corresponding to the target point by combining the relative position relationship and the vehicle model, the following steps are also included:

[0109] When acquiring new target point information, S300 first queries the mapping table to obtain the corresponding conversion point;

[0110] S400: If there is no matching item, add the calculated conversion point corresponding to the target point to the mapping table.

[0111] For logistics handling scenarios, the pickup area and unloading area in the warehouse are generally fixed, that is, the points on the map in the scene are generally fixed. When multiple unmanned forklifts are operating in the warehouse, for example, when a task is received to go to target point A to perform a cargo delivery task, the corresponding conversion point that adapts to the mechanical deviation of the unmanned forklift is calculated according to the technical solution disclosed in the embodiment of the present invention, and this target point and the corresponding conversion point are saved. When the same task information is received later, the computing power occupation and resource consumption caused by repeated calculations are avoided. By querying the mapping table, the conversion point corresponding to the target point can be directly determined, which is more efficient.

[0112] Embodiment 3

[0113] According to another aspect of the present invention, a control system for consistency of multiple unmanned forklifts is provided, the system comprising:

[0114] A storage unit, used to store the program of the steps of the consistency control method for multiple unmanned forklifts described in any embodiment of the present invention, so that the data acquisition unit, the control unit, and the processing unit can call and execute it in a timely manner;

[0115] A control unit, used to control the dynamic calibration plate to move within a predetermined range when performing external parameter calibration on the laser radar configured on the top of each unmanned forklift using the dynamic calibration plate; and control each unmanned forklift to navigate to a corresponding conversion point of the target point;

[0116] A data acquisition unit is used to collect laser point cloud data of a dynamic calibration plate when performing external parameter calibration on a laser radar configured on the top of each unmanned forklift using the dynamic calibration plate; and to collect point cloud data of a target area scanned by the laser radar when determining an actual control center offset parameter of each unmanned forklift according to a reference point cloud of a target area scanned by each unmanned forklift and a vehicle body model;

[0117] The processing unit is used to calculate the laser radar extrinsic calibration parameters based on the dynamic calibration plate point cloud data collected by the laser radar and the known geometric features of the dynamic calibration plate; calculate the actual control center offset parameters of each unmanned forklift according to the reference point cloud of the target area scanned by each unmanned forklift and the vehicle model, and calculate the relative position relationship between each unmanned forklift laser radar and the actual control center according to the actual control center offset parameters; calculate the conversion point corresponding to the target point according to the relative position relationship, target point information and the vehicle model.

[0118] Embodiment 4

[0119] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for controlling consistency of multiple unmanned forklifts as described in any embodiment of the present invention when executed.

[0120] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is only limited by the claims and their full scope and equivalents. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

[0121] Those skilled in the art can understand that, in addition to implementing the system, device, unit and its various modules provided by the present invention in a purely computer-readable program code, it is entirely possible to implement the same program in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers and embedded microcontrollers by logically programming the method steps. Therefore, the system, device and its various modules provided by the present invention can be considered as a hardware component, and the modules included therein for implementing various programs can also be regarded as structures within the hardware component; the modules for implementing various functions can also be regarded as both software programs for implementing the method and structures within the hardware component.

[0122] In addition, all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, and the program is stored in a storage medium, including a number of instructions to enable a single-chip microcomputer, a chip or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0123] In addition, various implementation modes of the embodiments of the present invention may be arbitrarily combined, and as long as they do not violate the concept of the embodiments of the present invention, they should also be regarded as the contents disclosed by the embodiments of the present invention.

[0124] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0125] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for controlling the consistency of multiple unmanned forklifts, characterized in that: include: Use a dynamic calibration plate to calibrate the external parameters of the laser radar configured on the top of each unmanned forklift; After control calibration, each unmanned forklift scans the target area to obtain the reference point cloud, determines the actual control center offset parameters based on the vehicle model, and determines the relative position relationship between the laser radar and the actual control center; The target point information is obtained, and the conversion point corresponding to the target point is calculated in combination with the relative position relationship and the vehicle body model, and each unmanned forklift is controlled to navigate to the conversion point.

2. The method according to claim 1, characterized in that Use the dynamic calibration board to calibrate the external parameters of the laser radar configured on the top of each unmanned forklift, including: The dynamic calibration plate includes N reflective marking points (N is a positive integer greater than 2), and the reflective marking points are arranged according to a predetermined geometric pattern; The dynamic calibration plate is scanned by the laser radar, and the external parameters of the laser radar are calculated based on the scanning result and the known position relationship of the reflection mark point.

3. The method according to claim 2, characterized in that Scanning the dynamic calibration plate by the laser radar, and calculating the external parameters of the laser radar based on the scanning result and the known position relationship of the reflection mark point, including: Controlling the dynamic calibration plate to move within a predetermined range, and during the movement, continuously collecting point cloud data of the dynamic calibration plate through the laser radar; Extracting geometric features of the dynamic calibration plate from the laser radar point cloud data; Using a nonlinear optimization algorithm, the external parameters of the laser radar are calculated based on the collected point cloud data and the known geometric features of the dynamic calibration plate.

4. The method according to claim 1, characterized in that: After the control calibration, each unmanned forklift scans the target area to obtain a reference point cloud, determines the actual control center offset parameter in combination with the vehicle body model, and determines the relative position relationship between the laser radar and the actual control center, including: Extracting feature points representing the contour of the fork arm of the unmanned forklift from the reference point cloud; A straight line fitting is performed on the characteristic points of the fork arm profile; and an offset parameter of the actual control center relative to the theoretical control center is calculated based on the fitting result and the vehicle body model.

5. The method according to claim 4, characterized in that The acquiring target point information and calculating the conversion point corresponding to the target point in combination with the relative position relationship and the vehicle body model includes: Determine the reference position point of each unmanned forklift laser radar according to the relative position relationship between each unmanned forklift laser radar and the actual control center; According to the reference position point of each unmanned forklift laser radar and the vehicle body model, the conversion point corresponding to the target point is determined.

6. The method according to claim 5, characterized in that Determining a conversion point corresponding to the target point according to the reference position point of each unmanned forklift laser radar and the vehicle body model includes: Determine, according to the constraints of the laser radar in the vehicle model and the theoretical control center in the vehicle model, the reference position information of the theoretical control center in the vehicle model when the laser radar is at the reference position point; The conversion point corresponding to the target point is determined according to the reference position information of the theoretical control center.

7. The method according to claim 6, characterized in that Determining the conversion point corresponding to the target point according to the reference position point of each unmanned forklift laser radar and the vehicle body model, further comprising: Determine, based on the laser radar extrinsic calibration parameters, a reference position point of the laser radar in the vehicle body coordinate system when the actual control center of the vehicle body is at the target point; Based on the vehicle model, calculating the theoretical control center position point of the vehicle model in the laser radar coordinate system; The theoretical control center position point of the vehicle model in the laser radar coordinate system is converted to the global map coordinate system to obtain a conversion point corresponding to the target point.

8. The method according to any one of claims 1 to 7, characterized in that: Establish a mapping table of target points and conversion points; Before obtaining the target point information and calculating the conversion point corresponding to the target point in combination with the relative position relationship and the vehicle model, the method further includes: When acquiring new target point information, the mapping table is preferentially queried to acquire the corresponding conversion point; If there is no target point match, the calculated conversion point corresponding to the target point is added to the mapping table.

9. A control system for consistency of multiple unmanned forklifts, characterized in that: The system comprises: A storage unit, used to store a program including the method steps as claimed in any one of claims 1 to 8, so that the data acquisition unit, the control unit, and the processing unit can call and execute it in a timely manner; A control unit, used to control the dynamic calibration board to move within a predetermined range when performing external parameter calibration on the laser radar configured on the top of each unmanned forklift using the dynamic calibration board; and control each unmanned forklift to navigate to a conversion point corresponding to the target point; A data acquisition unit is used to collect laser point cloud data of a dynamic calibration plate when performing external parameter calibration on a laser radar configured on the top of each unmanned forklift using the dynamic calibration plate; and to collect reference point cloud data of a target area scanned by the laser radar when determining the actual control center offset parameters of each unmanned forklift according to the reference point cloud scanned by each unmanned forklift and the vehicle body model; The processing unit is used to calculate the laser radar extrinsic calibration parameters based on the dynamic calibration plate point cloud data collected by the laser radar and the known geometric features of the dynamic calibration plate; calculate the actual control center offset parameters of each unmanned forklift according to the reference point cloud of the target area scanned by each unmanned forklift and the vehicle model, and calculate the relative position relationship between each unmanned forklift laser radar and the actual control center according to the actual control center offset parameters; calculate the conversion point corresponding to the target point according to the relative position relationship, target point information and the vehicle model.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the control method for consistency of multiple unmanned forklifts according to any one of claims 1 to 8 when executed.

Images