Mobile platform loading method and system of fork type unmanned vehicle and medium

Abstract

The invention discloses a mobile platform loading method and system of a fork type unmanned vehicle and a medium. The method comprises the steps that a loading starting signal of a WMS system is acquired; in response to the loading starting signal, controlling a space information scanning device to scan a mobile loading platform area to obtain original data; obtaining external parameters of the spatial information scanning device, and converting the original data into a predefined map coordinate system based on the external parameters; based on the converted data, extracting a loadable area of the mobile loading platform; based on the loadable area and predefined cargo size and placement gap parameters, automatically planning a placement storage location position of the cargo; planning a driving path leading to the position based on the parking space placement position; and the driving path is issued to the fork type unmanned vehicle so that the fork type unmanned vehicle can drive along the path and complete cargo loading. Full-process and self-adaptive automatic loading is achieved, and the flexible adaptability to different vehicle types and cargo sizes and the operation efficiency are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent logistics, in particular to a forklift unmanned vehicle mobile platform loading method, system and medium. BACKGROUND

[0002] Under the background of intelligence and automatic driving, forklift unmanned vehicles have been widely used in warehouse, logistics and other fields, and have achieved mature solutions. In particular, in the process of product off-line to warehouse transfer, the application of forklift unmanned vehicles has been quite mature. However, in the automatic mobile platform loading link, it is still a key bottleneck of automatic operation in the whole process. The popularization and improvement of automatic mobile platform loading are of great significance to improve the digitization and intelligence of the factory.

[0003] Traditional mobile platform loading is mainly completed by manual operation, which has many problems. First of all, manual operation cannot realize automatic inventory management of goods, which is easy to cause loss or error of goods information. Secondly, the placement quality of goods on the mobile platform depends largely on the experience of forklift drivers, which may lead to poor precision and consistency of goods placement. In addition, inexperienced forklift drivers may affect the speed and efficiency of goods turnover due to low operating efficiency.

[0004] Some existing forklift unmanned vehicles rely on global map operation after map updating in the automatic loading process. However, the effect of map stitching directly affects the loading precision. For some scenes with high requirements on placement precision, manual confirmation of map stitching quality may be required, which leads to the lack of full-automatic operation. Therefore, how to improve the precision of map stitching and how to ensure that the unmanned vehicle can realize full-automatic operation in the loading process have become urgent problems to be solved. SUMMARY

[0005] The purpose of the present application is to overcome the shortcomings of the prior art. In order to achieve the above purpose, a forklift unmanned vehicle mobile platform loading method, system and medium are used to solve the problems raised in the background technology.

[0006] A forklift unmanned vehicle mobile platform loading method, comprising the following steps: Step S1, obtaining a start loading signal of a WMS system; Step S2, in response to the start loading signal, controlling a space information scanning device to scan the mobile loading platform area to obtain raw data; Step S3, obtaining the extrinsic parameters of the space information scanning device, and converting the raw data into a predefined map coordinate system based on the extrinsic parameters; Step S4, based on the converted data, extracting the loadable area of the mobile loading platform; Step S5, automatically planning a placement storage location of the goods based on the loadable area and predefined goods size and placement gap parameters; Step S6, planning a driving path to the placement storage location based on the placement storage location; Step S7, issuing the driving path to the forkless unmanned vehicle, so that the forkless unmanned vehicle drives along the path and completes the goods loading.

[0007] As a further scheme of the present application: the start loading signal in step S1 includes a mobile loading platform parking area ID and goods size information, wherein the goods size information at least includes length, width and height of the goods.

[0008] As a further scheme of the present application: the specific steps in step S2 include: controlling the spatial information scanning device to rotate and scan at a preset angular velocity, and receiving raw data composed of motor angle and sensor data in real time.

[0009] As a further scheme of the present application: the external parameters of the spatial information scanning device in step S3 are obtained by a calibration method; The calibration method includes: presetting a marker of a known coordinate in a map coordinate system, obtaining the coordinate of the marker in the spatial information scanning device coordinate system by scanning, and then calculating a rotation matrix and a translation matrix from the spatial information scanning device coordinate system to the map coordinate system.

[0010] As a further scheme of the present application: in step S4, a hybrid region growing algorithm is used to extract plane point clouds representing the loadable area from the converted point cloud data.

[0011] As a further scheme of the present application: the specific steps in step S5 include: calculating the boundary corner points of the loadable area by a boundary fitting algorithm; determining the direction and size of the loadable area according to the boundary corner points; combining the predefined goods size and placement gap parameters, calculating the number of columns of loadable goods, the number of totes per column, and the placement pose of each tote of goods.

[0012] As a further scheme of the present application: the specific steps in step S6 include: setting a connection point outside the loadable area; calculating a front point according to the storage location; based on the motion performance of the vehicle, inserting a transition point before the front point to generate a smooth dynamic path from the connection point to the storage location.

[0013] As a further aspect of the application: the specific steps in step S7 include: The fork unmanned vehicle sends a path request after driving to the connection point; In response to the request, the current unloaded storage location and its corresponding travel path are issued; After the fork unmanned vehicle travels to the storage location according to the received path and completes the delivery, it returns to the connection point along the original path.

[0014] The technical solution of the second aspect: a mobile platform loading method system using any of the above fork unmanned vehicles, comprising: WMS system for sending a start loading signal; Spatial information scanning device for scanning the mobile loading platform area; A computing module in communication with the WMS system and the spatial information scanning device for calculating the travel path; The fork unmanned vehicle is in communication with the computing module for receiving the travel path and performing the loading task.

[0015] The technical solution of the third aspect: a storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the method of any of the above.

[0016] Compared with the prior art, the application has the following technical effects: By using the above technical solution, the original point cloud data of the loading platform is obtained through a specially designed spatial scanning device after receiving the instruction of the WMS system; the data is converted into a unified map coordinate system by using the pre-calibrated external parameters; the loadable area is intelligently extracted from the converted data by a point cloud processing algorithm; the optimal storage location of the goods is automatically calculated in the area according to the known size of the goods and the gap requirement; then the travel path of the unmanned vehicle is dynamically planned according to the storage location; finally, the path is issued to the fork unmanned vehicle to guide it to complete the precise and autonomous loading task. A full-automatic loading solution from environment perception, pose calculation to path planning is realized, which completely eliminates the dependence on fixed pre-set paths and fixed storage location mode, significantly improves the intelligent level and flexible adaptability of the loading task, can efficiently cope with different specifications of trucks, mobile platforms and various sizes of goods, greatly reduces manual intervention, and improves the efficiency and accuracy of logistics loading. BRIEF DESCRIPTION OF DRAWINGS

[0017] The specific embodiments of the application will be described in detail below with reference to the accompanying drawings: Figure 1 The steps of the mobile platform loading method of the disclosed embodiments are shown in the figure. Figure 2 This is a schematic diagram of the warehouse location pose according to an embodiment of this application. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please refer to Figure 1 In this embodiment of the invention, a method for loading a mobile platform for a forklift-type unmanned vehicle includes the following steps: Step S1: Obtain the start loading signal of the WMS system; In this embodiment, the start loading signal in step S1 includes the mobile loading platform parking area ID and cargo size information, wherein the cargo size information includes at least the length, width, and height of the cargo.

[0020] In the specific implementation steps, the WMS system sends a loading start signal to the computing device through the agreed network protocol. The WMS system and the computing device are located in the same local area network, and both are set with fixed IPs in the local area network. The content sent includes, but is not limited to, the agreed mobile loading platform parking area ID, the dimensions of the loaded goods, etc. The dimensions of the loaded goods include the length L, width W, and height H of the entire pallet of goods.

[0021] Step S2: In response to the start loading signal, control the spatial information scanning device to scan the mobile loading platform area to obtain raw data; In this embodiment, step S2 specifically includes the following steps: The spatial information scanning device is controlled to perform rotational scanning at a preset angular velocity, such as 1 degree / second, and to receive raw data composed of motor angles and sensor data in real time. The spatial information scanning device includes, but is not limited to, a single radar gimbal, a multi-radar scanning device, and a camera.

[0022] In the specific implementation steps, the computing device sends a start scanning signal to the spatial information scanning device corresponding to the area ID received from the WMS, based on the area ID received. The spatial information scanning device is designed as a motor with a motion angle of 0-180 degrees, which carries a single-line lidar rotating through a structural component. The radar's scanning sector is 270 degrees, and the motor's rotational angular velocity is 1 degree per second. The spatial information scanning device sends the motor angle and sensor data to the computing device in real time.

[0023] Step S3: Obtain the external parameters of the spatial information scanning device, and convert the raw data into a predefined map coordinate system based on the external parameters; In this embodiment, the external parameters of the spatial information scanning device in step S3 are obtained through a calibration method; The calibration method includes: In the map coordinate system, a landmark with known coordinates within the map coordinate system is preset. The coordinates of the landmark in the spatial information scanning device coordinate system are obtained by scanning. Then, the rotation matrix and translation matrix from the spatial information scanning device coordinate system to the map coordinate system are calculated.

[0024] The specific implementation steps are as follows: After acquiring the scanned point cloud for an entire cycle, the computing device obtains the extrinsic parameters of the spatial information scanning device in the map coordinate system and transforms the point cloud into the map coordinate system. The extrinsic parameter calibration method for the spatial information scanning device is described as follows: Place markers with known coordinates within the parking area designed for the mobile, loadable platform. , , , , ,in , , , To form a quadrilateral in the map coordinate system, the following conditions must be met. , Within the map coordinate system Two points on the axis, and exist The position on the axis is greater than exist On the axis, the distance between the two is not less than 10 meters. Let the position of the aforementioned marker in the map coordinate system be any location within the quadrilateral described above. , , , ,

[0025] Furthermore, it is required to use a level to place the aforementioned markers on the same horizontal plane, and drive the spatial information scanning device to scan and obtain the coordinates of the aforementioned markers within the spatial information scanning device. , , , , The method for solving the extrinsic parameters of the spatial information scanning device in the map coordinate system is as follows: the coordinate system transformation matrix is ​​represented as a homogeneous matrix. ,in Let be the rotation matrix and be the translation matrix. Given... , The angle of the ray formed in the map coordinate system is 0°, according to , By calculating the angle θ within the spatial information scanning device, the rotation relationship can be obtained. From this, we can obtain New coordinates after rotation ,Depend on and The translation relationship between two points is obtained by deriving the translation relationship. .

[0026] Step S4: Based on the converted data, extract the loadable area of ​​the mobile loading platform; In this embodiment, step S4 uses a hybrid region growing algorithm to extract planar point clouds representing loadable regions from the converted point cloud data.

[0027] Specifically, the transformation relationship between the gimbal and the map coordinate system, solved in step S3, is used to transform the scanned point cloud into the map coordinate system after one scan. It is known that the loading area in the map coordinate system consists of one or more planes. For this structured scene, a hybrid region growing algorithm is proposed to extract the planar point cloud within the loading area.

[0028] Step S5: Based on the loadable area and predefined cargo size and placement gap parameters, automatically plan the placement location of the cargo. In this embodiment, step S5 specifically includes the following steps: The boundary corner points of the loadable region are calculated using a boundary fitting algorithm; The orientation and size of the loadable area are determined based on the boundary corner points; Based on the predefined cargo dimensions and placement gap parameters, the number of columns that can be loaded with cargo, the number of pallets per column, and the placement orientation of each pallet of cargo are calculated.

[0029] Specifically, such as Figure 2 As shown in the figure, the diagram is a schematic diagram of the warehouse location pose; Based on the results of step S4, the four corner points of the loading area are calculated using the boundary fitting method. , , , ,in , Given two points on the longer side, the angle of that longer side can be calculated. , , Let the two points on the other longer side be denoted as and the angle of that longer side be denoted as . , , Using two points on the shorter side of the vehicle facing forward, the width of the vehicle on that side can be determined. , , Using two points on the shorter side of the vehicle towards the rear, the width of that side can be determined. It reads parameters stored in the medium, such as the size of the cargo to be loaded, the distance from the edge of the carriage after loading, and the distance between two adjacent pallets, and calculates the number of cargo columns that can be loaded, the number of pallets that can be loaded in each column, and the position of each pallet.

[0030] Step S6: Based on the location of the storage location, plan the driving route to that location; In this embodiment, step S6 specifically includes the following steps: A docking point is set outside the loadable area, specifically a fixed point outside the docking area of ​​the mobile loading platform area is set as the docking point, i.e. the starting point; Calculate the preceding point based on the location of the storage location; Based on the vehicle's motion performance, a transition point is inserted before the forward point to generate a smooth dynamic path from the docking point to the storage location.

[0031] Specifically, the steps of dynamic path planning are as follows: It is a dynamic path planning algorithm that does not rely on preset paths. Traditional methods obtain new paths by offsetting preset paths. This method requires drawing a preset road network based on the standard parking posture of trucks during implementation. If the size of the goods to be loaded changes, or the truck size changes, or the quantity of goods to be loaded changes, the preset paths cannot fully cover business needs. The newly designed dynamic path method adaptively calculates the number of storage locations to be planned and the dynamic paths that can reach these storage locations based on truck size, the size of the goods to be loaded, etc.

[0032] The dynamic path calculation method is as follows: a connection point is set on each of the left and right sides outside the loading area, which serves as the connection point between the dynamic loading road network and the pre-set transportation road network.

[0033] Based on the location pose calculated in step S5, the calculation system reads the distance between the storage location and the previous point stored in the medium. The pose of the forward point is calculated. Based on the longitudinal distance between the forward point and the docking point, a point preceding the forward point is inserted. The pose of the point preceding the forward point is set appropriately according to the vehicle performance. , The axial distance is 90 degrees apart.

[0034] Step S7: The driving path is sent to the forklift truck so that the forklift truck can travel along the path and complete the loading of goods.

[0035] In this embodiment, step S7 specifically includes the following steps: The forklift-type unmanned vehicle sends a path request after arriving at the docking point; In response to the request, the location of the currently unloaded storage location and its corresponding driving path are sent out; The forklift truck travels to the storage location according to the received path, completes the unloading, and then returns to the docking point along the original path.

[0036] In the specific implementation steps, after receiving the loading task sent by the dispatcher, the forklift truck will carry the goods to the docking point. After arriving at the docking point, it will send a dynamic path request signal to the system. After receiving the signal, the system will send the position of the unloaded storage location and the associated path according to the dynamic planning storage location loading completion mark. The forklift truck will then walk to the target location according to the path to complete the unloading and return to the docking point according to the reverse path of entering the storage location to complete the loading of the goods.

[0037] The second aspect of the technical solution: A system employing a mobile platform loading method for a forklift-type unmanned vehicle as described in any of the above claims, comprising: The WMS system is used to send the start loading signal; Spatial information scanning device, used to scan the area of ​​the mobile loading platform; The calculation module is communicatively connected to the WMS system and the spatial information scanning device, and is used to calculate the driving path; The forklift-type unmanned vehicle is communicatively connected to the computing module and is used to receive the driving path and perform loading operations.

[0038] The third aspect of the technical solution: a storage medium storing a computer program thereon, wherein the computer program, when executed by a processor, implements the method described in any of the above.

[0039] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention. The scope of the invention is defined by the appended claims and their equivalents, all of which should be included within the scope of protection of the invention.

Claims

1. A method for loading a mobile platform of a forked unmanned vehicle, characterized in that, The method comprises the following steps: Step S1, obtaining a start loading signal of a WMS system; Step S2, in response to the start loading signal, controlling a spatial information scanning device to scan a mobile loading platform area to obtain raw data; Step S3, obtaining extrinsic parameters of the spatial information scanning device, and converting the raw data into a predefined map coordinate system based on the extrinsic parameters; Step S4, based on the converted data, extracting a loadable area of the mobile loading platform; Step S5, based on the loadable area and predefined cargo size and placement gap parameters, automatically planning a placement storage location of the cargo; Step S6, based on the placement storage location, planning a driving path to the location; Step S7, issuing the driving path to a fork unmanned vehicle, so that the fork unmanned vehicle drives along the path and completes cargo loading.

2. The method of claim 1, wherein, The start loading signal in step S1 includes a mobile loading platform parking area ID and cargo size information, wherein the cargo size information at least includes length, width and height of the cargo.

3. The method of claim 1, wherein, The specific steps in step S2 include: The spatial information scanning device is controlled to rotate and scan at a preset angular velocity, and raw data composed of motor angle and sensor data is received in real time.

4. The method of claim 1, wherein, The extrinsic parameters of the spatial information scanning device in step S3 are obtained by a calibration method; The calibration method comprises: A marker of a known coordinate in the map coordinate system is preset, the coordinates of the marker in the spatial information scanning device coordinate system are obtained by scanning, and a rotation matrix and a translation matrix from the spatial information scanning device coordinate system to the map coordinate system are calculated.

5. The method of claim 1, wherein, In step S4, a hybrid region growing algorithm is used to extract plane point cloud representing the loadable area from the converted point cloud data.

6. The method of claim 1, wherein, The specific steps in step S5 include: The boundary corner points of the loadable area are calculated by a boundary fitting algorithm; The direction and size of the loadable area are determined according to the boundary corner points; The number of columns, the number of totes per column and the placement pose of each tote cargo of the loadable cargo are calculated in combination with the predefined cargo size and placement gap parameters.

7. The method of claim 1, wherein the forked unmanned vehicle is a forked unmanned aerial vehicle. The specific steps in step S6 include: A connection point is set outside the loadable area; A front point is calculated according to the storage location; Based on the motion performance of the vehicle, a transition point is inserted before the front point to generate a smooth dynamic path from the connection point to the storage location.

8. The method of claim 1, wherein, The specific steps in step S7 include: The fork unmanned vehicle sends a path request after driving to the connection point; In response to the request, the currently unloaded storage location and its corresponding driving path are issued; After the fork unmanned vehicle drives to the storage location according to the received path to complete the cargo placement, it returns to the connection point along the original path.

9. A system for loading a mobile platform with a forked unmanned vehicle according to any one of claims 1 to 8, characterized in that, It comprises: A WMS system for sending a start loading signal; A spatial information scanning device for scanning a mobile loading platform area; A computing module in communication connection with the WMS system and the spatial information scanning device for calculating a driving path; A fork unmanned vehicle in communication connection with the computing module for receiving the driving path and performing loading operation.

10. A storage medium having stored thereon a computer program, characterized in that The computer program, which is executed by a processor, implements the method as claimed in any of claims 1 to 8.

Images