Novel automatic loading and unloading system of unmanned forklift in van environment

Abstract

The invention relates to the technical field of automatic loading and unloading of logistics forklifts, in particular to a novel automatic loading and unloading system of an unmanned forklift in a van environment, and the system comprises a positioning map binding module which is used for binding platform areas where a plurality of trucks are parked on a positioning map; and the task issuing module is configured to receive a task issued by a driver through a pad system when a truck arrives and transmit the task to a warehouse management system (WMS), and the WMS issues a cargo taking and placing task of a corresponding map point position according to a platform number. And the visual servo system adopting the solid-state laser radar provides high-precision and high-stability navigation capability for the AGV.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic loading and unloading of logistics forklifts, and in particular to a new automatic loading and unloading system for unmanned forklifts in a van environment. Background Art

[0002] Driven by the booming global economy and the rapid expansion of the e-commerce industry, the logistics industry is booming at an unprecedented rate. This trend has not only greatly promoted the circulation of goods, but also put forward higher requirements for improving logistics efficiency and reducing costs. As a representative of automated logistics equipment, the automatic guided vehicle (AGV) has been widely used in the logistics industry due to its high efficiency and precision, especially in the automatic loading and unloading tasks of trucks, showing great potential.

[0003] However, the automatic loading and unloading of trucks faces a series of complex challenges in actual operation. As a common vehicle type in logistics transportation, vans have various surface conditions, ranging from flat and smooth surfaces to complex environments with uneven surfaces and stains. This diversity places extremely high demands on the positioning accuracy and navigation capabilities of unmanned forklifts, requiring the system to be able to quickly adapt to different compartment environments to ensure the stability and accuracy of loading and unloading operations.

[0004] In addition, the cargo clearance inside the truck is small and the loading and unloading space is limited, which undoubtedly increases the difficulty of unmanned forklifts to operate accurately in a small space. Unmanned forklifts not only need to have high-precision positioning capabilities, but also need to flexibly adjust their posture in a limited space to ensure that the cargo can be safely and accurately loaded and unloaded to the designated location.

[0005] What is even more difficult is that the automatic loading and unloading tasks in the van environment often require the AGV to be able to go deep into the narrow compartment to pick up and place goods. This requirement not only tests the mechanical design and motion control capabilities of the unmanned forklift, but also puts higher demands on its perception and decision-making capabilities in complex environments. How to ensure the safe and stable operation of the unmanned forklift in the narrow compartment while ensuring operational efficiency has become a key issue that needs to be solved in the current logistics automation field. Summary of the invention

[0006] (I) Purpose of the invention

[0007] In view of this, the purpose of the present invention is to propose a new unmanned forklift automatic loading and unloading system in a van environment, which solves the problems raised in the above background.

[0008] (II) Technical solution

[0009] A new unmanned forklift automatic loading and unloading system in a van environment, the system includes:

[0010] A positioning map binding module is used to bind the platform area where multiple trucks stop to the positioning map;

[0011] The task issuing module is configured to receive the task issued by the driver through the pad system when the truck arrives, and pass the task to the warehouse management system (WMS). The WMS system issues the pick-up and delivery tasks of the corresponding map points according to the platform number;

[0012] An automated guided vehicle (AGV) navigation module that uses a reflector-based positioning system to get the AGV to the dock where the truck is parked;

[0013] A 3D point cloud template generation module receives the length, width and height information of the van input by the user in the offline stage, and generates a 3D point cloud of the van as a template point cloud based on the information, which is used to plan the storage locations where cargo can be loaded;

[0014] The real-time scanning and registration module is configured to scan the vehicle body point cloud in real time when the forklift arrives at the platform, and register it with the offline template point cloud to obtain the real-time navigation posture of the forklift;

[0015] The precise navigation module receives the current target point position obtained in real time and transmits the position to the control system to guide the forklift to the storage location of the van;

[0016] The vehicle-mounted vision task execution module performs precise vehicle-mounted vision tasks after the forklift arrives at the storage location to place the goods more accurately at the storage location;

[0017] The template point cloud update module updates the template point cloud according to the changes inside the carriage each time a loading or unloading task is completed;

[0018] The visual navigation module is used to perform registration operations to obtain navigation poses and update the template point cloud after each delivery to improve navigation efficiency;

[0019] A pallet posture recognition module is configured to scan the pallet using a solid-state laser radar when performing a pickup task inside a van, segment the pallet point cloud using a point cloud segmentation algorithm, fit a plane to obtain a pallet angle and a pallet center coordinate, and send the pallet angle and the pallet center coordinate to a control system;

[0020] The cargo placement posture adjustment module uses a solid-state laser radar to scan the cargo placement location when performing cargo placement tasks inside the van, and post-processes the point cloud of the compartment at the cargo placement location to determine the left and right offsets that need to be further adjusted by the forklift to avoid scratching the cargo.

[0021] Preferably, the vehicle-mounted visual task execution module adopts a high-precision visual recognition algorithm to achieve accurate placement of goods.

[0022] Preferably, the real-time scanning and registration module adopts an advanced point cloud registration algorithm to achieve fast and accurate real-time navigation posture acquisition.

[0023] Preferably, the three-dimensional point cloud template generation module adopts an advanced discretization algorithm to generate a high-precision three-dimensional point cloud template of the van.

[0024] Preferably, the pallet posture recognition module and the cargo placement posture adjustment module both use solid-state laser radar as the main sensing device to improve the stability and accuracy of the system.

[0025] A new method for automatic loading and unloading of unmanned forklifts in a van environment, the method comprising the following steps:

[0026] 1). Task issuance: WMS (Warehouse Management System) issues loading tasks for specific platforms according to demand;

[0027] 2). Task execution and monitoring: The unmanned forklift receives the task and starts to execute it, checking whether the loading task is completed: if the task is completed, the whole process ends; if the task is not completed, continue to execute the following steps;

[0028] 3) Real-time scanning and registration: The laser radar scans the car point cloud in real time, and registers the scanned car point cloud with the offline template point cloud (such as using the ICP algorithm) to obtain the precise position and posture of the current target point;

[0029] 4). Posture transmission and navigation: The real-time target point posture is transmitted to the control system; the control system controls the unmanned forklift to navigate to the specified location according to the target point posture;

[0030] 5) Accurate cargo placement and visual tasks: When the unmanned forklift arrives at the target point, it performs accurate cargo placement visual tasks, such as using the on-board camera for visual recognition to ensure cargo placement accuracy and cargo safety;

[0031] 6). Template point cloud update: After each delivery, the template point cloud is updated according to the new state of the interior of the carriage to reflect the latest carriage structure;

[0032] 7) Loop judgment: After updating the template point cloud, return to step 2 and recheck whether the loading task is completed. If the task is still not completed, repeat steps 3 to 6 until WMS confirms that the loading task is completed.

[0033] It can be seen from the above technical solutions that this application has the following beneficial effects:

[0034] 1. This invention uses a solid-state laser radar visual servo system to navigate the AGV in a van scenario, and the solid-state laser radar visual servo system provides the AGV with high-precision and high-stability navigation capabilities. In the complex and restricted environment of a van, the solid-state laser radar can perceive the surrounding environment in real time and provide the AGV with accurate navigation information to ensure that it can travel accurately and efficiently in the narrow compartment. The application of this technology not only improves the navigation accuracy of the AGV, but also greatly enhances its adaptability in complex environments, providing strong support for the automation and intelligent development of the logistics industry.

[0035] 2. The invention is based on the solid-state laser radar scanning of the van, which can correct the deviation of the truck's parking position, so that during the truck's parking process, due to various factors (such as driver operation, uneven road surface), there may be deviations in the truck's parking position. By scanning the van with a solid-state laser radar, the truck's parking position information can be obtained in real time, and the parking deviation can be accurately corrected accordingly. This function not only improves the accuracy of the truck's parking, but also ensures that the AGV can smoothly enter the interior of the car when loading and unloading goods, avoiding operation delays or safety accidents caused by parking deviations.

[0036] 3. The invention establishes templates offline and updates them dynamically after the task is completed. The solid-state laser radar does not need long-term integration in the online stage, which is more efficient. By establishing a three-dimensional point cloud template of the van offline and dynamically updating it according to the real-time scanning data after the task is completed, it can ensure that the AGV always has high-precision reference information during the navigation and positioning process. At the same time, the solid-state laser radar does not need long-term integration in the online stage, which greatly improves the response speed and operating efficiency of the system. This technical feature enables AGV to complete loading and unloading tasks more quickly and accurately in complex and changing environments, providing a strong guarantee for the rapid development of the logistics industry. In addition, the dynamically updated template can also adapt to changes in different vehicle models and cargo loading conditions, improving the flexibility and adaptability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a schematic diagram of the automatic loading process of a van of the present invention;

[0038] Figure 2 It is a schematic diagram of the automatic unloading process of a van truck according to the present invention. DETAILED DESCRIPTION

[0039] The following description is merely exemplary in nature and is not intended to limit the present disclosure, application, and use. It should be understood that in all of these figures, the same or similar reference numerals indicate the same or similar parts and features. The various drawings only schematically represent the concepts and principles of the embodiments of the present disclosure, and do not necessarily show the specific dimensions and proportions of the various embodiments of the present disclosure. Specific parts in specific drawings may be exaggerated to illustrate the relevant details or structures of the embodiments of the present disclosure.

[0040] See also Figure 1-2 , an embodiment provided by the present invention:

[0041] A new unmanned forklift automatic loading and unloading system in a van environment, the system includes:

[0042] A positioning map binding module is used to bind the platform area where multiple trucks stop to the positioning map;

[0043] The task issuing module is configured to receive the task issued by the driver through the pad system when the truck arrives, and pass the task to the warehouse management system (WMS). The WMS system issues the pick-up and delivery tasks of the corresponding map points according to the platform number;

[0044] An automated guided vehicle (AGV) navigation module that uses a reflector-based positioning system to get the AGV to the dock where the truck is parked;

[0045] The 3D point cloud template generation module receives the length, width and height information of the truck input by the user in the offline stage, and generates a 3D point cloud of the van as a template point cloud based on the information, which is used to plan the storage locations where cargo can be loaded;

[0046] The real-time scanning and registration module is configured to scan the vehicle body point cloud in real time when the forklift arrives at the platform, and register it with the offline template point cloud to obtain the real-time navigation posture of the forklift;

[0047] The precise navigation module receives the current target point position obtained in real time and transmits the position to the control system to guide the forklift to the storage location of the van;

[0048] The vehicle-mounted vision task execution module performs precise vehicle-mounted vision tasks after the forklift arrives at the storage location to place the goods more accurately at the storage location;

[0049] The template point cloud update module updates the template point cloud according to the changes inside the carriage each time a loading or unloading task is completed;

[0050] The visual navigation module is used to perform registration operations to obtain navigation poses and update the template point cloud after each delivery to improve navigation efficiency;

[0051] A pallet posture recognition module is configured to scan the pallet using a solid-state laser radar when performing a pickup task inside a van, segment the pallet point cloud using a point cloud segmentation algorithm, fit a plane to obtain the pallet angle and the pallet center coordinates, and send the pallet angle and the pallet center coordinates to a control system;

[0052] The cargo placement posture adjustment module uses a solid-state laser radar to scan the cargo placement location when performing cargo placement tasks inside the van, and post-processes the point cloud of the compartment at the cargo placement location to determine the left and right offsets that need to be further adjusted by the forklift to avoid scratching the cargo.

[0053] Furthermore, the on-board visual task execution module adopts a high-precision visual recognition algorithm to achieve accurate placement of goods. The on-board visual task execution module adopts a high-precision visual recognition algorithm to achieve accurate identification and positioning of goods, ensuring accurate placement of goods during loading and unloading. This not only improves the accuracy of loading and unloading operations, but also reduces losses caused by misplacement or damage of goods, and improves overall logistics efficiency.

[0054] Furthermore, the real-time scanning and registration module uses advanced point cloud registration algorithms to achieve fast and accurate real-time navigation posture acquisition. The real-time scanning and registration module uses advanced point cloud registration algorithms to quickly and accurately acquire the real-time navigation posture of the unmanned forklift in the carriage. This technical feature enables the unmanned forklift to navigate flexibly in complex and changeable carriage environments, improving the adaptability and flexibility of the system.

[0055] Furthermore, the 3D point cloud template generation module uses advanced discretization algorithms to generate high-precision 3D point cloud templates for vans. This template provides an accurate reference for the navigation and positioning of unmanned forklifts, further improving the navigation accuracy and operating efficiency of the system.

[0056] Furthermore, the pallet posture recognition module and the cargo placement posture adjustment module both use solid-state laser radar as the main sensing device to improve the stability and accuracy of the system. The solid-state laser radar has the advantages of high stability, accurate measurement, and strong anti-interference ability, which can ensure that the unmanned forklift can stably and accurately perform the pallet posture recognition and cargo placement posture adjustment tasks in complex environments. This technical feature not only improves the reliability and stability of the system, but also provides strong support for the widespread application of unmanned forklifts.

[0057] A new method for automatic loading and unloading of unmanned forklifts in a van environment, the method comprising the following steps:

[0058] 1). Task issuance: WMS (Warehouse Management System) issues loading tasks for specific platforms according to demand;

[0059] 2). Task execution and monitoring: The unmanned forklift receives the task and starts to execute it, checking whether the loading task is completed: if the task is completed, the whole process ends; if the task is not completed, continue to execute the following steps;

[0060] 3) Real-time scanning and registration: The laser radar scans the car point cloud in real time, and registers the scanned car point cloud with the offline template point cloud (such as using the ICP algorithm) to obtain the precise position and posture of the current target point;

[0061] 4). Posture transmission and navigation: The real-time target point posture is transmitted to the control system; the control system controls the unmanned forklift to navigate to the specified location according to the target point posture;

[0062] 5) Accurate cargo placement and visual tasks: When the unmanned forklift arrives at the target point, it performs accurate cargo placement visual tasks, such as using the on-board camera for visual recognition to ensure cargo placement accuracy and cargo safety;

[0063] 6). Template point cloud update: After each delivery, the template point cloud is updated according to the new state of the interior of the carriage to reflect the latest carriage structure;

[0064] 7). Loop judgment: After updating the template point cloud, return to step 2 and recheck whether the loading task is completed. If the task is still not completed, repeat steps 3 to 6 until WMS confirms that the loading task is completed.

[0065] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A new unmanned forklift automatic loading and unloading system in a van environment, characterized by: The system includes: A positioning map binding module is used to bind the platform area where multiple trucks stop to the positioning map; The task issuing module is configured to receive the task issued by the driver through the pad system when the truck arrives, and pass the task to the warehouse management system (WMS). The WMS system issues the pick-up and delivery tasks of the corresponding map points according to the platform number; An automated guided vehicle (AGV) navigation module that uses a reflector-based positioning system to get the AGV to the dock where the truck is parked; A 3D point cloud template generation module receives the length, width and height information of the van input by the user in the offline stage, and generates a 3D point cloud of the van as a template point cloud based on the information, which is used to plan the storage locations where cargo can be loaded; The real-time scanning and registration module is configured to scan the vehicle body point cloud in real time when the forklift arrives at the platform, and register it with the offline template point cloud to obtain the real-time navigation posture of the forklift; The precise navigation module receives the current target point position obtained in real time and transmits the position to the control system to guide the forklift to the storage location of the van; The vehicle-mounted vision task execution module performs precise vehicle-mounted vision tasks after the forklift arrives at the storage location to place the goods more accurately at the storage location; The template point cloud update module updates the template point cloud according to the changes inside the carriage each time a loading or unloading task is completed; The visual navigation module is used to perform registration operations to obtain navigation poses and update the template point cloud after each delivery to improve navigation efficiency; A pallet posture recognition module is configured to scan the pallet using a solid-state laser radar when performing a pickup task inside a van, segment the pallet point cloud using a point cloud segmentation algorithm, fit a plane to obtain a pallet angle and a pallet center coordinate, and send the pallet angle and the pallet center coordinate to a control system; The cargo placement posture adjustment module uses a solid-state laser radar to scan the cargo placement location when performing cargo placement tasks inside the van, and post-processes the point cloud of the compartment at the cargo placement location to determine the left and right offsets that need to be further adjusted by the forklift to avoid scratching the cargo.

2. A new unmanned forklift automatic loading and unloading system in a van environment according to claim 1, characterized in that: The vehicle-mounted visual task execution module adopts a high-precision visual recognition algorithm to achieve accurate placement of goods.

3. A new unmanned forklift automatic loading and unloading system in a van environment according to claim 1, characterized in that: The real-time scanning and registration module adopts advanced point cloud registration algorithm to achieve fast and accurate real-time navigation posture acquisition.

4. A new unmanned forklift automatic loading and unloading system in a van environment according to claim 1, characterized in that: The three-dimensional point cloud template generation module adopts an advanced discretization algorithm to generate a high-precision three-dimensional point cloud template of a van.

5. According to claim 1, a new unmanned forklift automatic loading and unloading system in a van environment is characterized by: The pallet posture recognition module and the cargo placement posture adjustment module both use solid-state laser radar as the main sensing device to improve the stability and accuracy of the system.

6. A new method for automatic loading and unloading of unmanned forklifts in a van environment, characterized by: The method comprises the following steps: 1). Task issuance: WMS (Warehouse Management System) issues loading tasks for specific platforms according to demand; 2). Task execution and monitoring: The unmanned forklift receives the task and starts to execute it, checking whether the loading task is completed: if the task is completed, the whole process ends; if the task is not completed, continue to execute the following steps; 3) Real-time scanning and registration: The laser radar scans the car point cloud in real time, and registers the scanned car point cloud with the offline template point cloud (such as using the ICP algorithm) to obtain the precise position and posture of the current target point; 4). Posture transmission and navigation: The real-time target point posture is transmitted to the control system; the control system controls the unmanned forklift to navigate to the specified location according to the target point posture; 5) Accurate cargo placement and visual tasks: When the unmanned forklift arrives at the target point, it performs accurate cargo placement visual tasks, such as using the on-board camera for visual recognition to ensure cargo placement accuracy and cargo safety; 6). Template point cloud update: After each delivery, the template point cloud is updated according to the new state of the interior of the carriage to reflect the latest carriage structure; 7) Loop judgment: After updating the template point cloud, return to step 2 and recheck whether the loading task is completed. If the task is still not completed, repeat steps 3 to 6 until WMS confirms that the loading task is completed.

Images