Carrier stacking method, device and equipment

By obtaining the target point cloud of the symmetrical support components of the vehicle and establishing a grid map, determining the positioning posture, the alignment problem during the robot vehicle stacking process is solved, and a safe and accurate stacking effect is achieved.

CN118906054AActive Publication Date: 2024-11-08HANGZHOU HIKROBOT TECH CO LTD
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202411018182.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-26
Publication Date
2024-11-08
Estimated Expiration
2044-07-26

AI Technical Summary

Technical Problem

In the prior art, robots have a risk of dumping due to misalignment of vehicles during vehicle stacking, poor safety and lack effective stacking solutions.

Method used

By obtaining the target point cloud of the symmetrical support components of the vehicle, establishing a grid map under the robot coordinate system, determining the position of the vehicle, and controlling the robot to stack the vehicle onto the storage position based on the target position to ensure alignment and safety .

Benefits of technology

The safety and accuracy of vehicle stacking is achieved, the risk of dumping is avoided, and the reliability of the stacking process is improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118906054A_ABST
    Figure CN118906054A_ABST
Patent Text Reader

Abstract

The invention provides a carrier stacking method, device and equipment. The method comprises the steps that a first target point cloud of a symmetrical supporting component of a first carrier and a second target point cloud of a symmetrical supporting component of a second carrier are obtained; establishing a grid map under a robot coordinate system; based on the position of the first target point cloud in the grid map, selecting a first target grid, in the grid map, of the symmetrical supporting part of the first carrier; based on the position of the second target point cloud in the grid map, selecting a second target grid, in the grid map, of the symmetrical supporting part of the second carrier; determining a first pose of the first carrier in the robot coordinate system based on the first target grid, and determining a second pose of the second carrier in the robot coordinate system based on the second target grid; determining a target pose of the robot based on the first pose and the second pose; and controlling the robot to stack the first carrier on the second carrier based on the target pose. According to the scheme, the carrier on the robot is completely aligned with the carrier at the storage position.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of robot control technology, and in particular to a carrier stacking method, device and equipment. Background Art

[0002] In recent years, various types of robots (such as autonomous mobile robots, etc.) have developed rapidly in terms of technology and market. Robots are mechanical devices that automatically perform work and rely on their own power and control capabilities to achieve various functions. Robots can accept human commands, run pre-programmed programs, and act according to strategies formulated by artificial intelligence. For example, users use manual remote controls to control robots to perform related operations. For example, the manual remote control sends operation commands to the robot wirelessly. After receiving the operation command, the robot executes the operation specified by the operation command to complete the related functions.

[0003] With the rapid development of robot technology, robots are increasingly used in logistics, warehousing, factory production, etc. For example, robots can be used to transport carriers and achieve carrier stacking, which means that the robot stacks the carrier on the robot onto the carrier in the storage location.

[0004] However, there is no effective stacking solution for stacking the carriers on the robot onto the carriers in the storage location. For example, when the robot is lowering a carrier, if the carrier on the robot is not completely aligned with the carrier in the storage location, the carrier cannot be lowered, and there is a risk of tipping over when lowering the carrier, which is less safe. Summary of the invention

[0005] The present application provides a carrier stacking method, wherein a first carrier is on a robot and a second carrier is on a storage location, and the method comprises:

[0006] Acquire a first target point cloud of a symmetrical support component of a first carrier and a second target point cloud of a symmetrical support component of a second carrier, wherein the symmetrical support component is a contact component when two carriers are stacked;

[0007] Establishing a grid map in the robot coordinate system, wherein the grid map includes a plurality of grids;

[0008] Based on the position of the first target point cloud in the grid map, a first target grid corresponding to the symmetrical support component of the first vehicle in the grid map is selected from the multiple grids; based on the position of the second target point cloud in the grid map, a second target grid corresponding to the symmetrical support component of the second vehicle in the grid map is selected from the multiple grids;

[0009] Determine a first pose of the first vehicle in a robot coordinate system based on the first target grid, and determine a second pose of the second vehicle in a robot coordinate system based on the second target grid;

[0010] Determining a target posture of the robot based on the first posture and the second posture;

[0011] The robot is controlled based on the target posture to stack the first carrier onto the second carrier.

[0012] The present application provides a carrier stacking device, wherein a first carrier is disposed on a robot and a second carrier is disposed on a storage location, and the device comprises:

[0013] An acquisition module, used to acquire a first target point cloud of a symmetrical support component of a first carrier and a second target point cloud of a symmetrical support component of a second carrier, wherein the symmetrical support component is a contact component when two carriers are stacked;

[0014] A selection module is used to establish a grid map in the robot coordinate system, wherein the grid map includes a plurality of grids; based on the position of the first target point cloud in the grid map, select a first target grid corresponding to the symmetrical support component of the first vehicle in the grid map from the plurality of grids; based on the position of the second target point cloud in the grid map, select a second target grid corresponding to the symmetrical support component of the second vehicle in the grid map from the plurality of grids;

[0015] A determination module, configured to determine a first pose of the first vehicle in a robot coordinate system based on the first target grid, determine a second pose of the second vehicle in a robot coordinate system based on the second target grid; and determine a target pose of the robot based on the first pose and the second pose;

[0016] A control module is used to control the robot to stack the first carrier onto the second carrier based on the target posture.

[0017] The present application provides an electronic device, comprising: a processor and a machine-readable storage medium, wherein the machine-readable storage medium stores machine-executable instructions that can be executed by the processor; the processor is used to execute the machine-executable instructions to implement the carrier stacking method of the above example of the present application.

[0018] The present application provides a carrier stacking system, the system comprising a dispatching device and a robot, the robot being equipped with a radar, the robot being equipped with a first carrier, and the storage being equipped with a second carrier; wherein the radar is arranged on a toothing of the robot and moves up and down with the toothing; or the radar is arranged on a moving part of the robot and moves up and down with the moving part;

[0019] The scheduling device is used to obtain the configured stacking preparation position and the configured target height, and send the stacking preparation position and the target height to the robot;

[0020] The robot is used to move to the stacking preparation position and raise the height of the radar to the target height; collect original point clouds through the radar, and obtain a first target point cloud of the symmetrical support component of the first carrier and a second target point cloud of the symmetrical support component of the second carrier based on the original point cloud, where the symmetrical support component is a contact component when two carriers are stacked;

[0021] A grid map is established in the robot coordinate system, wherein the grid map includes multiple grids; based on the position of the first target point cloud in the grid map, a first target grid corresponding to the symmetrical support component of the first vehicle in the grid map is selected from the multiple grids; based on the position of the second target point cloud in the grid map, a second target grid corresponding to the symmetrical support component of the second vehicle in the grid map is selected from the multiple grids; a target posture of the robot is determined based on the first target grid and the second target grid; and based on the target posture, the robot is controlled to stack the first vehicle onto the second vehicle.

[0022] It can be seen from the above technical solutions that in the embodiment of the present application, based on the first target point cloud of the symmetrical support components of the first carrier (carrier on the robot) and the second target point cloud of the symmetrical support components of the second carrier (carrier on the storage location), the first posture of the first carrier in the robot coordinate system and the second posture of the second carrier in the robot coordinate system are determined, the target posture of the robot is determined based on the first posture and the second posture, and the robot is controlled to stack the first carrier on the second carrier based on the target posture. By determining the target posture of the robot in the world coordinate system, the first carrier on the robot is completely aligned with the second carrier in the storage location. When the robot lowers the carrier, the first carrier is aligned with the second carrier, there is no risk of tipping over, the safety is high, and the carrier stacking can be effectively completed, and the carrier on the robot can be accurately stacked on the carrier in the storage location. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 is a schematic flow chart of a carrier stacking method in one embodiment of the present application;

[0024] Figure 2 is a schematic flow chart of a carrier stacking method in one embodiment of the present application;

[0025] Figure 3 It is a schematic diagram of the positional relationship between the stacking preparation position and the storage position in one embodiment of the present application;

[0026] Figure 4 It is a schematic diagram of dividing a grid map into four areas in one implementation of the present application;

[0027] Figure 5 is a schematic structural diagram of a carrier stacking device in one embodiment of the present application;

[0028] Figure 6 It is a hardware structure diagram of an electronic device in one embodiment of the present application. DETAILED DESCRIPTION

[0029] A carrier stacking method is proposed in an embodiment of the present application, where there is a first carrier on the robot (for the convenience of distinction, the carrier on the robot is referred to as the first carrier), there is a second carrier on the storage location (for the convenience of distinction, the carrier on the storage location is referred to as the second carrier), and the first carrier needs to be stacked onto the second carrier.

[0030] See also Figure 1 FIG. 1 is a flow chart of the carrier stacking method, which may include:

[0031] Step 101: Acquire a first target point cloud of a symmetrical support component of a first carrier and a second target point cloud of a symmetrical support component of a second carrier, wherein the symmetrical support component is a contact component when two carriers are stacked.

[0032] Step 102: Establish a grid map in the robot coordinate system, the grid map including a plurality of grids. Based on the position of the first target point cloud in the grid map, select a first target grid corresponding to the symmetrical support component of the first vehicle in the grid map from the plurality of grids; based on the position of the second target point cloud in the grid map, select a second target grid corresponding to the symmetrical support component of the second vehicle in the grid map from the plurality of grids.

[0033] For example, if the robot is a forklift, the robot coordinate system may be a forklift coordinate system, and the forklift coordinate system is used as an example later. Of course, the present embodiment is not limited to the forklift coordinate system.

[0034] Step 103: determine a first pose of the first vehicle in the robot coordinate system based on the first target grid, and determine a second pose of the second vehicle in the robot coordinate system based on the second target grid.

[0035] Step 104: Determine the target posture of the robot based on the first posture and the second posture.

[0036] Step 105: Control the robot to stack the first carrier onto the second carrier based on the target posture.

[0037] Exemplarily, obtaining the first target point cloud of the symmetrical support component of the first carrier and the second target point cloud of the symmetrical support component of the second carrier may include, but is not limited to: after controlling the robot to move to the configured stacking preparation position and raising the height of the radar to the configured target height, collecting the original point cloud through the radar, the original point cloud includes the point cloud of the first carrier and the point cloud of the second carrier; wherein the radar is deployed on the robot. For example, the radar can be deployed on the robot's toothing (taking the robot as a forklift as an example, the forklift includes a toothing, and the carrier stacking is completed by moving the toothing up and down), and the radar can move up and down with the toothing. Alternatively, the radar is deployed on the moving part of the robot and moves up and down with the moving part. For example, a slider and a slide rail can be deployed on the robot, and the radar can be deployed on the slider, so that the radar can be controlled to move up and down when the slider moves up and down on the slide rail.

[0038] Based on the position and posture of the radar in the robot coordinate system, determine the rotation and translation matrix from the radar coordinate system to the robot coordinate system; transform the original point cloud to the robot coordinate system through the rotation and translation matrix to obtain the reference point cloud in the robot coordinate system; select the first reference point cloud within the ROI range of the first vehicle from the reference point cloud, and perform semantic segmentation on each position point in the first reference point cloud to obtain semantic information; select the second reference point cloud within the ROI range of the second vehicle from the reference point cloud, and perform semantic segmentation on each position point in the second reference point cloud to obtain semantic information; wherein the semantic information indicates whether the position point is a symmetrical support component or not; select the position points belonging to the symmetrical support components from the first reference point cloud to form a first target point cloud, and select the position points belonging to the symmetrical support components from the second reference point cloud to form a second target point cloud.

[0039] Exemplarily, before selecting the first reference point cloud within the ROI (Region Of Interest) range of the first vehicle from the reference point cloud, the ROI range of the first vehicle can also be determined based on the rough pose of the first vehicle in the robot coordinate system, the length dimension of the first vehicle, the width dimension of the first vehicle, the target height of the radar, the length error tolerance value, the width error tolerance value, and the height error tolerance value. In addition, the rough pose of the second vehicle in the robot coordinate system can be determined based on the center point pose of the storage location and the pose of the robot in the world coordinate system; the ROI range of the second vehicle can be determined based on the rough pose, the length dimension of the second vehicle, the width dimension of the second vehicle, the target height of the radar, the length error tolerance value, and the width error tolerance value.

[0040] Exemplarily, based on the position of the first target point cloud in the grid map, selecting the first target grid corresponding to the symmetrical support component of the first carrier in the grid map from multiple grids may include but is not limited to: dividing the grid map into multiple regions (the number of regions matches the number of contact components when stacked), each region includes multiple grids. Based on this, for each region, the first score value of the grid is determined based on the distance between the grid in the region and the position of the first target point cloud in the grid map, and multiple first candidate grids are selected from the region based on the first score value of each grid; and the first target grid is determined from the multiple first candidate grids in the region. In this way, for each region, the region corresponds to a first target grid.

[0041] Exemplarily, determining the first score value of the grid based on the distance between the grid in the region and the position of the first target point cloud in the grid map may include but is not limited to: for each position point of the first target point cloud, determining the grid coordinates of the position point in the grid map; for each grid in the region, determining the score value corresponding to the position point in the grid based on the distance between the grid and the grid coordinates; wherein, the smaller the distance, the larger the score value. The first score value of each grid in the region is counted, and the first score value may be the sum of the score values ​​corresponding to each position point of the first target point cloud in the grid.

[0042] Furthermore, for each area of ​​the grid map, multiple first candidate grids can be selected from the area based on the first score value of each grid in the area, and the first target grids can be traversed from the multiple first candidate grids in the area so that the first target grids of the multiple areas meet the constraint conditions.

[0043] Exemplarily, multiple first candidate grids are selected in order of first score values ​​from high to low, and the distance between different first candidate grids is greater than a threshold. The multiple regions may include a left front region, a right front region, a left rear region, and a right rear region. On this basis, when determining the first target grid from the multiple first candidate grids in the region, each region corresponds to a first target grid, and the first target grids corresponding to the multiple regions meet the constraint conditions. The first target grids corresponding to the multiple regions meet the constraint conditions, which may include but are not limited to: the distance between the first target grid Pi in the left front region and the first target grid Pj in the right front region meets the vehicle size constraint; the distance between the first target grid Pj and the first target grid Pk in the right rear region meets the vehicle size constraint, and the absolute value of the difference between the first angle and 90 degrees is less than the threshold; wherein the first angle is the angle between the first target grid Pi, the first target grid Pj, and the first target grid Pk. The distance between the first target grid Pi and the first target grid Pl in the left rear area satisfies the vehicle size constraint, and the absolute value of the difference between the second angle and 90 degrees is less than the threshold; wherein the second angle is the angle between the first target grid Pj, the first target grid Pi, and the first target grid Pl.

[0044] Exemplarily, based on the position of the second target point cloud in the grid map, selecting the second target grid corresponding to the symmetrical support component of the second vehicle in the grid map from multiple grids may include: dividing the grid map into multiple regions, each region including multiple grids. Based on this, for each region, the second score value of the grid is determined based on the distance between the grid in the region and the position of the second target point cloud in the grid map, and multiple second candidate grids are selected from the region based on the second score value of each grid; and the second target grid is determined from the multiple second candidate grids in the region, that is, each region corresponds to a first target grid.

[0045] Exemplarily, determining the second score of the grid based on the distance between the grid in the region and the position of the second target point cloud in the grid map may include but is not limited to: for each position point of the second target point cloud, determining the grid coordinates of the position point in the grid map; for each grid in the region, determining the score corresponding to the position point in the grid based on the distance between the grid and the grid coordinates; wherein the smaller the distance, the larger the score. The second score of each grid in the region is counted, and the second score may be the sum of the scores corresponding to each position point of the second target point cloud in the grid.

[0046] Furthermore, for each area of ​​the grid map, multiple second candidate grids can be selected from the area based on the second score value of each grid in the area, and second target grids can be traversed from the multiple second candidate grids in the area so that the second target grids in multiple areas meet the constraint conditions.

[0047] Exemplarily, determining the target posture of the robot based on the first posture and the second posture may include but is not limited to: converting the first posture into a first rotation and translation matrix, converting the second posture into a second rotation and translation matrix, and converting the posture of the robot in the world coordinate system into a third rotation and translation matrix; determining the fourth rotation and translation matrix of the first vehicle in the world coordinate system based on the first rotation and translation matrix and the third rotation and translation matrix; determining the fifth rotation and translation matrix of the second vehicle in the world coordinate system based on the second rotation and translation matrix and the third rotation and translation matrix; determining the sixth rotation and translation matrix of the robot in the first vehicle coordinate system based on the fourth rotation and translation matrix and the third rotation and translation matrix; determining the target rotation and translation matrix of the target posture of the robot in the world coordinate system based on the sixth rotation and translation matrix and the fifth rotation and translation matrix, and converting the target rotation and translation matrix into the target posture of the robot in the world coordinate system.

[0048] Exemplarily, controlling the robot to stack the first carrier onto the second carrier based on the target posture may include but is not limited to: controlling the robot to move to the target position based on the target posture; after the robot moves to the target position, obtaining a third target point cloud of the symmetrical supporting components of the first carrier and a fourth target point cloud of the symmetrical supporting components of the second carrier; determining whether there is a stacking deviation between the first carrier and the second carrier based on the third target point cloud and the fourth target point cloud; if not, controlling the robot to stack the first carrier onto the second carrier; if so, returning to execute the operation of obtaining the first target point cloud of the symmetrical supporting components of the first carrier and the second target point cloud of the symmetrical supporting components of the second carrier.

[0049] Exemplarily, determining whether there is a stacking deviation between the first vehicle and the second vehicle based on the third target point cloud and the fourth target point cloud may include but is not limited to: selecting a third target grid from the grid map based on the third target point cloud, and selecting a fourth target grid from the grid map based on the fourth target point cloud; determining a third pose of the first vehicle in the robot coordinate system based on the third target grid, and determining a fourth pose of the second vehicle in the robot coordinate system based on the fourth target grid; if the deviation between the third target grid and the first target grid is less than a threshold, the deviation between the fourth target grid and the second target grid is less than a threshold, the deviation between the third pose and the first pose is less than a threshold, and the deviation between the fourth pose and the second pose is less than a threshold, then it is determined that there is no stacking deviation between the first vehicle and the second vehicle.

[0050] It can be seen from the above technical solutions that in the embodiment of the present application, based on the first target point cloud of the symmetrical support components of the first carrier (carrier on the robot) and the second target point cloud of the symmetrical support components of the second carrier (carrier on the storage location), the first posture of the first carrier in the robot coordinate system and the second posture of the second carrier in the robot coordinate system are determined, the target posture of the robot is determined based on the first posture and the second posture, and the robot is controlled to stack the first carrier on the second carrier based on the target posture. By determining the target posture of the robot in the world coordinate system, the first carrier on the robot is completely aligned with the second carrier in the storage location. When the robot lowers the carrier, the first carrier is aligned with the second carrier, there is no risk of tipping over, the safety is high, and the carrier stacking can be effectively completed, and the carrier on the robot can be accurately stacked on the carrier in the storage location.

[0051] The above technical solutions of the embodiments of the present application are described below in combination with specific application scenarios.

[0052] Carrier stacking means stacking the carrier on the robot onto the carrier in the storage location. For the sake of distinction, the carrier on the robot is called the first carrier, and the carrier on the storage location is called the second carrier. In order to achieve the carrier stacking function, the size of the first carrier on the robot is the same as the size of the second carrier in the storage location, such as the length dimension of the first carrier is the same as the length dimension of the second carrier, the width dimension of the first carrier is the same as the width dimension of the second carrier, and the height dimension of the first carrier is the same as or different from the height dimension of the second carrier. In this way, the first carrier and the second carrier can be aligned and placed.

[0053] The robot in this embodiment can be a forklift or other types of robots that can realize carrier stacking. The forklift will be used as an example for the following description. The storage location in this embodiment is a storage location for placing the carrier. The carrier in this embodiment can be a device for storing materials, such as a shelf.

[0054] See also Figure 2 FIG. 1 is a flow chart of a carrier stacking method, which may include:

[0055] Step 201: Control the forklift to move to the stacking preparation position and lift the radar to the target height.

[0056] Exemplarily, the carrier stacking method can be applied to a forklift itself, that is, the forklift itself executes the carrier stacking method. Alternatively, the carrier stacking method can also be applied to a forklift control device, which can be called a dispatching platform, and the dispatching platform executes the carrier stacking method.

[0057] For example, a stacking preparation position may be pre-configured, and the stacking preparation position is any position close to the storage position, and there is no restriction on the stacking preparation position. Figure 3 , which is a schematic diagram of the positional relationship between the stacking preparation position and the storage position. After the forklift is controlled to move to the stacking preparation position, the forklift approaches the storage position. When the forklift moves to the stacking preparation position, the point cloud collected by the radar may include the point cloud of the first carrier (i.e., the carrier on the gear insert) and the point cloud of the second carrier (i.e., the carrier on the storage position).

[0058] Exemplarily, when the forklift moves to the stacking preparation position, the radar height can be raised to a target height, which can be pre-configured. There is no restriction on this target height. As long as the radar is raised to the target height, the point cloud collected by the radar includes the point cloud of the first vehicle and the point cloud of the second vehicle.

[0059] To summarize, the stacking preparation position and radar target height can be pre-configured. After controlling the forklift to move to the stacking preparation position and raising the radar height to the target height, the point cloud collected by the radar includes the point cloud of the first vehicle and the point cloud of the second vehicle. There is no restriction on the stacking preparation position and radar target height.

[0060] For example, the radar can be deployed on the forklift's tooth (such as fixed below the tooth), and the radar can move up and down with the tooth. Alternatively, the radar is deployed on the moving parts of the robot and moves up and down with the moving parts. For example, a slider and a slide rail can be deployed on the robot, and the radar can be deployed on the slider, so that when the slider moves up and down on the slide rail, the radar can be controlled to move up and down.

[0061] Step 202: collect original point clouds through radar, and pre-process the original point clouds to obtain reference point clouds.

[0062] Exemplarily, the following steps may be used to implement point cloud preprocessing to obtain a reference point cloud:

[0063] Step 2021: collect original point cloud by radar (the point cloud collected by radar is called original point cloud).

[0064] For example, the radar can be used to obtain a set number of point clouds and cache them, and these point clouds are recorded as original point clouds. For example, after the forklift moves to the stacking preparation position and the radar (the radar can be a laser radar or other types of radar) is raised to the target height, the radar can be used to collect the original point cloud within the field of view, and the original point cloud includes the point cloud of the first vehicle and the point cloud of the second vehicle.

[0065] Step 2022: Based on the position and orientation of the radar in the forklift coordinate system (i.e., the position and orientation of the radar relative to the forklift coordinate system), determine the rotation and translation matrix from the radar coordinate system to the forklift coordinate system.

[0066] For example, assuming the pose It is used to represent the pose of the A coordinate system in the B coordinate system, that is, the pose of the radar coordinate system in the forklift coordinate system (such as the position and direction of the radar relative to the forklift coordinate system). The following formula (1) is used to determine the rotation and translation matrix from the radar coordinate system to the forklift coordinate system:

[0067]

[0068] Step 2023: The original point cloud is transformed into the forklift coordinate system by using the rotation and translation matrix to obtain a reference point cloud in the forklift coordinate system (ie, the point cloud in the forklift coordinate system can be called a reference point cloud).

[0069] For example, the original point cloud is collected by radar, so the original point cloud is a point cloud in the radar coordinate system. After obtaining the rotation and translation matrix from the radar coordinate system to the forklift coordinate system, the original point cloud in the radar coordinate system can be converted into a reference point cloud in the forklift coordinate system based on the rotation and translation matrix.

[0070] At this point, the introduction of step 202 is completed, and the subsequent step 203 can be continued.

[0071] Step 203: Perform semantic segmentation based on the reference point cloud to obtain semantic information.

[0072] Exemplarily, the following steps can be used to implement semantic segmentation and obtain semantic information:

[0073] Step 2031: Determine the ROI range of the first vehicle on the forklift.

[0074] For example, the rough position and posture of the first vehicle in the forklift coordinate system can be determined. For example, since the positional relationship between the forklift's tine and the forklift's center point is known, when the forklift coordinate system is established with the forklift's center point as the origin, the position and posture of the tine in the forklift coordinate system can be determined based on the positional relationship.

[0075] Since the first carrier is placed on the forklift's tine, the position and posture of the first carrier in the forklift coordinate system can be determined based on the position and posture of the tine in the forklift coordinate system. Of course, this position and posture is not accurate and can be called a rough position and posture of the first carrier in the forklift coordinate system. For example, the position and posture of the tine in the forklift coordinate system is used as the rough position and posture of the first carrier in the forklift coordinate system, or the position and posture of the tine in the forklift coordinate system is optimized to obtain the rough position and posture of the first carrier in the forklift coordinate system.

[0076] In one possible implementation, the ROI range of the first vehicle can be determined based on the rough posture of the first vehicle in the forklift coordinate system, the length dimension of the first vehicle, the width dimension of the first vehicle, the target height of the radar, the length error tolerance value, the width error tolerance value and the height error tolerance value.

[0077] For example, when a forklift is stacking from the length direction of the carrier, the following formula is used to determine the ROI range of the first carrier: zmin=h_reco-h_pole, xmax=h_reco.

[0078] For example, when a forklift is stacking from the width of the carrier, the following formula is used to determine the ROI range of the first carrier: zmin=h_reco-h_pole, zmax=h_reco.

[0079] In the above formula, xmin and xmax represent the horizontal minimum and horizontal maximum values ​​of the ROI range of the first carrier, ymin and ymax represent the vertical minimum and vertical maximum values ​​of the ROI range of the first carrier, and zmin and zmax represent the vertical minimum and vertical maximum values ​​of the ROI range of the first carrier.

[0080] x and y represent the rough position of the first vehicle in the forklift coordinate system, W represents the width dimension of the first vehicle, tol_W represents the width error tolerance value, which can be configured based on experience, L represents the length dimension of the first vehicle, tol_L represents the length error tolerance value, which can be configured based on experience. h_reco represents the target height of the radar, that is, the current height of the radar, and h_pole represents the height error tolerance value, which can be configured based on experience.

[0081] Step 2032: Determine the ROI range of the second carrier on the storage location.

[0082] Exemplarily, the rough pose of the second vehicle in the forklift coordinate system can be determined. For example, the rough pose of the second vehicle in the forklift coordinate system can be determined based on the pose of the center point of the storage location and the pose of the forklift in the world coordinate system. For example, the pose of the center point of the storage location represents the pose of the center point of the storage location in the world coordinate system, and the pose of the center point of the storage location can be determined based on the global map, that is, the center point of the storage location is found from the global map, and the position of the center point of the storage location corresponds to the pose of the center point of the storage location. The pose of the forklift in the world coordinate system represents the pose of the forklift coordinate system in the world coordinate system.

[0083] Obviously, based on the position of the center point of the storage location in the world coordinate system and the position of the forklift coordinate system in the world coordinate system, the position of the center point of the storage location in the forklift coordinate system can be obtained.

[0084] Since the second vehicle is placed on the storage location, the position and posture of the second vehicle in the forklift coordinate system can be determined based on the position and posture of the center point of the storage location in the forklift coordinate system. Of course, this position and posture is not accurate and can be called the rough position and posture of the second vehicle in the forklift coordinate system. For example, the position and posture of the center point of the storage location in the forklift coordinate system is used as the rough position and posture of the second vehicle in the forklift coordinate system, or the position and posture of the center point of the storage location in the forklift coordinate system is optimized to obtain the rough position and posture of the second vehicle in the forklift coordinate system.

[0085] In one possible implementation, the ROI range of the second vehicle can be determined based on the rough posture of the second vehicle in the forklift coordinate system, the length dimension of the second vehicle, the width dimension of the second vehicle, the target height of the radar, the length error tolerance value, and the width error tolerance value.

[0086] For example, when a forklift is stacking from the length direction of the carrier, the following formula is used to determine the ROI range of the second carrier: zmin′=0, zmax′=h_reco.

[0087] For example, when a forklift is stacking from the width of the carrier, the following formula is used to determine the ROI range of the second carrier: zmin′=0, zmax′=h_cero.

[0088] In the above formula, xmin′ and xmax′ represent the horizontal minimum and maximum values ​​of the ROI range of the second carrier, ymin′ and ymax′ represent the vertical minimum and maximum values ​​of the ROI range of the second carrier, and zmin′ and zmax′ represent the vertical minimum and maximum values ​​of the ROI range of the second carrier.

[0089] x′ and y′ can represent the rough position of the second vehicle in the forklift coordinate system, W can represent the width dimension of the second vehicle, tol_W can represent the width error tolerance value, which can be configured based on experience, L can represent the length dimension of the second vehicle, tol_L can represent the length error tolerance value, which can be configured based on experience. h_reco represents the target height of the radar, that is, the current height of the radar.

[0090] Step 2033: Select a first reference point cloud within the ROI range of the first vehicle from the reference point cloud, and select a second reference point cloud within the ROI range of the second vehicle from the reference point cloud.

[0091] Exemplarily, each position point in the reference point cloud can be traversed. If the position point is within the ROI of the first vehicle, the position point is added to the first reference point cloud. If the position point is within the ROI of the second vehicle, the position point is added to the second reference point cloud. If the position point is neither within the ROI of the first vehicle nor within the ROI of the second vehicle, the position point is ignored.

[0092] After performing the above processing on each position point in the reference point cloud, a first reference point cloud and a second reference point cloud can be obtained, and the first reference point cloud and the second reference point cloud both include a plurality of position points.

[0093] Step 2034: Perform semantic segmentation on each position point in the first reference point cloud to obtain semantic information, and perform semantic segmentation on each position point in the second reference point cloud to obtain semantic information. For each position point, the semantic information of the position point indicates whether the position point is a symmetrical support component or not a symmetrical support component.

[0094] In a possible implementation, a point cloud semantic segmentation model can be used to perform semantic segmentation to obtain semantic information, or other methods can be used to perform semantic segmentation to obtain semantic information. There is no limitation on this semantic segmentation method. The following uses a point cloud semantic segmentation model to perform semantic segmentation as an example.

[0095] A point cloud semantic segmentation model may be pre-trained, and the point cloud semantic segmentation model may be PointNet or the like, and the point cloud semantic segmentation model is used to identify whether each position point in the point cloud is a symmetrical supporting component.

[0096] On this basis, the first reference point cloud can be input into the point cloud semantic segmentation model, and the point cloud semantic segmentation model performs semantic segmentation on each position point in the first reference point cloud to obtain the semantic information of the position point. The semantic information indicates whether the position point is a symmetrical support component or not. The semantic information can also be called the semantic type of the position point. For example, a symmetrical support component can be a contact component when two carriers are stacked. The contact component can be a connecting component or another type of component as long as the carriers can be stacked. For example, the symmetrical support component of the first carrier is connected to the symmetrical support component of the second carrier, so that the first carrier is stacked on the second carrier. The semantic information can also be called the semantic type of the position point.

[0097] Similarly, the second reference point cloud can be input into the point cloud semantic segmentation model, which performs semantic segmentation on each position point in the second reference point cloud to obtain semantic information of the position point, which indicates whether the position point is a symmetrical supporting component or not.

[0098] At this point, the introduction of step 203 is completed, and the subsequent step 204 can be continued.

[0099] Step 204: Select a target point cloud of the symmetrical support component from the reference point cloud based on the semantic information. For example, select the position points belonging to the symmetrical support component from the first reference point cloud to form a first target point cloud, and select the position points belonging to the symmetrical support component from the second reference point cloud to form a second target point cloud.

[0100] Exemplarily, after obtaining the semantic information of each position point in the first reference point cloud, position points (i.e., three-dimensional points) belonging to the symmetrical support components can be selected from all position points in the first reference point cloud based on the semantic information, and these position points belonging to the symmetrical support components constitute the first target point cloud, that is, the first target point cloud is the point cloud of the symmetrical support components of the first vehicle. After obtaining the semantic information of each position point in the second reference point cloud, position points (i.e., three-dimensional points) belonging to the symmetrical support components can be selected from all position points in the second reference point cloud based on the semantic information, and these position points belonging to the symmetrical support components constitute the second target point cloud, that is, the second target point cloud is the point cloud of the symmetrical support components of the first vehicle.

[0101] Step 205: Select multiple candidate grids from all grids in the grid map based on the target point cloud.

[0102] Exemplarily, based on the target point cloud, multiple candidate grids can be selected using the following steps:

[0103] Step 2051: Create a grid map, such as a two-dimensional grid map, in a forklift coordinate system.

[0104] Exemplarily, the range of the grid map needs to be larger than the ranges of the first vehicle and the second vehicle, that is, the first vehicle needs to be located within the grid map, and the second vehicle needs to be located within the grid map.

[0105] For example, when a grid map is established in the forklift coordinate system, it is assumed that the coordinates of the origin of the grid map in the forklift coordinate system are The size of a single grid cell of the grid map is C.

[0106] Exemplarily, the grid map may include multiple grids, and the value of each grid of the grid map represents the score of the grid, and the score represents the score of the grid. When the grid map is initially established, the score of each grid of the grid map is initialized to 0 (or other values).

[0107] Step 2052: for each position point in the first target point cloud, determine the grid coordinates of the position point in the grid map, and the grid coordinates may correspond to a grid in the grid map.

[0108] For example, each position point in the first target point cloud can be traversed, and for the currently traversed position point The following formula (2) is used to determine the grid coordinates of the location point on the grid map:

[0109]

[0110] In the above formula, since the first target point cloud is the point cloud in the forklift coordinate system, It can represent the coordinates of the position point in the forklift coordinate system. It can represent the coordinates of the origin of the grid map in the forklift coordinate system, and C can represent the size of a single grid of the grid map.

[0111] Step 2053: After obtaining the grid coordinates of the location point in the grid map, for each grid in the grid map, based on the distance between the grid and the grid coordinates, determine the score value corresponding to the location point in the grid. For example, the smaller the distance between the grid and the grid coordinates, the larger the score value.

[0112] For example, assuming the grid coordinates Corresponding to grid Ci, for each grid in the grid map, based on the distance between the grid and grid Ci, determine the score value s corresponding to the position point in the grid, and the score value s can be greater than or equal to 0 and less than or equal to 1. For example, for grid Ci in the grid map, if the distance between the grid and grid Ci is 0, the score value s of the grid can be 1. For the grid in the grid map, if the distance between the grid and grid Ci is greater than or equal to Dmax, Dmax can be configured based on experience, then the score value s of the grid can be 0. For the grid in the grid map, if the distance between the grid and grid Ci is between 0 and Dmax, then the score value s of the grid is inversely proportional to the distance, for example, s=f(d), d represents the distance between the grid and grid Ci, f represents the functional relationship, which can be designed according to the actual application, such as the functional relationship f can be a linear decreasing relationship, there is no restriction on this, as long as s is inversely proportional to the distance d.

[0113] For example, an example of s=f(d) can be: s=1-d / Dmax, where d represents the distance between the grid and the grid Ci. Obviously, when d is Dmax, the score value s can be 0. Of course, when d is greater than Dmax, the score value s can also be determined to be 0. When d is, the score value s can be 1.

[0114] Step 2054: Count the first score value of each grid in the grid map. The first score value may be the sum of the score values ​​corresponding to each position point of the first target point cloud in the grid.

[0115] Exemplarily, for each grid in the grid map, after performing the above processing on each position point in the first target point cloud, the score value corresponding to the position point in the grid can be obtained, and the score value may be 1, the score value may be 0, or the score value may be a value between 0 and 1. Obviously, each position point corresponds to a score value in the grid. On this basis, the sum of the score values ​​corresponding to all the position points of the first target point cloud in the grid can be taken as the first score value of the grid.

[0116] In summary, the first score value of each grid in the grid map can be counted.

[0117] Step 2055: For each area of ​​the grid map, multiple first candidate grids can be selected from the area based on the first score value of each grid in the area. The multiple first candidate grids are selected in descending order of the first score values, and the distance between different first candidate grids is greater than a threshold.

[0118] For example, the grid map can be divided into multiple areas according to the theoretical center position of the vehicle (such as the center point of the grid map), and the number of the multiple areas is the same as the number of symmetrical support components of the vehicle. Assuming that there are 4 symmetrical support components, the grid map can be divided into 4 areas, namely the left front area (left front area), the right front area (right front area), the left rear area (left rear area), and the right rear area (right rear area), see Figure 4 As shown, it is a schematic diagram of dividing the grid map into four areas.

[0119] For the left front area, multiple first candidate grids can be selected from the left front area based on the first score value of each grid in the left front area. These multiple first candidate grids need to be selected in descending order of the first score values, and the distance between different first candidate grids can be greater than a threshold.

[0120] For example, several grids with high scores and mutual distances greater than a threshold are searched and recorded as first candidate grids (grids with high scores), and the first candidate grids are sorted from high to low according to the scores.

[0121] For example, suppose that M first candidate grids need to be selected, and each grid is traversed in order from high to low in terms of the first score value. For the currently traversed grid, first determine whether the distance between the grid and the selected first candidate grid is greater than the threshold. If not, abandon the grid and continue to traverse the next grid.

[0122] If so, the grid is taken as the selected first candidate grid, and it is determined whether the number of first candidate grids reaches M. If so, the traversal is stopped to obtain M first candidate grids. If not, the traversal continues to the next grid, and so on, until M first candidate grids in the left front area are obtained, or although M first candidate grids are not obtained, but the score value of the currently traversed grid is less than the threshold, the traversal is stopped.

[0123] For the right front region, multiple first candidate grids may be selected from the right front region based on the first score value of each grid in the right front region. For the left rear region, multiple first candidate grids may be selected from the left rear region based on the first score value of each grid in the left rear region. For the right rear region, multiple first candidate grids may be selected from the right rear region based on the first score value of each grid in the right rear region.

[0124] Step 2056: for each position point in the second target point cloud, determine the grid coordinates of the position point in the grid map, and the grid coordinates may correspond to a grid in the grid map.

[0125] Step 2057: After obtaining the grid coordinates of the location point in the grid map, for each grid in the grid map, determine the score value corresponding to the location point in the grid based on the distance between the grid and the grid coordinates. For example, the smaller the distance between the grid and the grid coordinates, the larger the score value.

[0126] Step 2058: Count the second score value of each grid in the grid map. The second score value may be the sum of the score values ​​corresponding to each position point of the second target point cloud in the grid.

[0127] Step 2059: For each area of ​​the grid map, multiple second candidate grids can be selected from the area based on the second score value of each grid in the area. The multiple second candidate grids are selected in descending order of the second score values, and the distance between different second candidate grids is greater than a threshold.

[0128] Exemplarily, steps 2056-2059 can refer to steps 2052-2055, which will not be repeated here.

[0129] At this point, the introduction of step 205 is completed, and the subsequent step 206 can be continued.

[0130] Step 206, select a target grid from multiple candidate grids. For example, traverse the first target grid from multiple first candidate grids in each area (such as the left front area, the right front area, the left rear area, and the right rear area) so that the first target grids in multiple areas meet the constraint conditions. Traverse the second target grid from multiple second candidate grids in each area so that the second target grids in multiple areas meet the constraint conditions. The first target grid represents the grid corresponding to the symmetrical support component of the first vehicle in the grid map, and the second target grid represents the grid corresponding to the symmetrical support component of the second vehicle in the grid map.

[0131] For example, a first target grid S11 is selected from a plurality of first candidate grids in the left front region, a first target grid S12 is selected from a plurality of first candidate grids in the right front region, a first target grid S13 is selected from a plurality of first candidate grids in the left rear region, and a first target grid S14 is selected from a plurality of first candidate grids in the right rear region, and the first target grid S11, the first target grid S12, the first target grid S13, and the first target grid S14 need to satisfy a constraint condition. In addition, a second target grid S21 is selected from a plurality of second candidate grids in the left front region, a second target grid S22 is selected from a plurality of second candidate grids in the right front region, a second target grid S23 is selected from a plurality of second candidate grids in the left rear region, and a second target grid S24 is selected from a plurality of second candidate grids in the right rear region, and the second target grid S21, the second target grid S22, the second target grid S23, and the second target grid S24 need to satisfy a constraint condition.

[0132] In a possible implementation, the first target grid of the left front area is recorded as Pi, the first target grid of the right front area is recorded as Pj, the first target grid of the right rear area is recorded as Pk, and the first target grid of the left rear area is recorded as Pl. On this basis, the first target grids of multiple areas meet the constraint conditions, which may include but are not limited to: the distance between the first target grid Pi and the first target grid Pj meets the vehicle size constraint. The distance between the first target grid Pj and the first target grid Pk meets the vehicle size constraint, and the absolute value of the difference between the first angle and 90 degrees is less than the threshold value, and the first angle is the angle between the first target grid Pi, the first target grid Pj, and the first target grid Pk. The distance between the first target grid Pi and the first target grid Pl meets the vehicle size constraint, and the absolute value of the difference between the second angle and 90 degrees is less than the threshold value, and the second angle is the angle between the first target grid Pj, the first target grid Pi, and the first target grid Pl.

[0133] The second target grid of the left front area is recorded as Pi', the second target grid of the right front area is recorded as Pj', the second target grid of the right rear area is recorded as Pk', and the second target grid of the left rear area is recorded as Pl'. On this basis, the second target grids of multiple areas meet the constraint conditions, which may include but are not limited to: the distance between the second target grid Pi' and the second target grid Pj' meets the vehicle size constraint. The distance between the second target grid Pj' and the second target grid Pk' meets the vehicle size constraint, and the absolute value of the difference between the second angle and 90 degrees is less than the threshold value, and the second angle is the angle between the second target grid Pi', the second target grid Pj', and the second target grid Pk'. The distance between the second target grid Pi' and the second target grid Pl' meets the vehicle size constraint, and the absolute value of the difference between the second angle and 90 degrees is less than the threshold value, and the second angle is the angle between the second target grid Pj', the second target grid Pi', and the second target grid Pl'.

[0134] In a possible implementation, the selection process is described by taking the implementation process of the first target grids of multiple regions satisfying the constraint conditions as an example, and the second target grids of multiple regions satisfying the constraint conditions are similar.

[0135] S2061: If all first candidate grids (grids with the highest score) in the left front region have been traversed, the loop is exited and the search ends. Otherwise, the next first candidate grid Pi is taken from the left front region in sequence.

[0136] S2062: If all first candidate grids (grids with the highest scores) in the right front region have been traversed, the process returns to S2061. Otherwise, a first candidate grid Pj is sequentially selected from the right front region.

[0137] S2063: Determine whether the distance between Pi and Pj satisfies the vehicle size constraint. If so, proceed to S2064. If not, return to S2062.

[0138] S2064: If all first candidate grids (grids with the highest scores) in the right rear region have been traversed, the process returns to S2062. Otherwise, a first candidate grid Pj is taken from the right rear region.

[0139] Determine whether the distance between Pj and Pk satisfies the vehicle size constraint, and determine whether the absolute value of the difference between the angle PiPjPk and 90 degrees is less than the set threshold. If both conditions are met, record Pk and proceed to S2065. If both conditions are not met, return to S2064.

[0140] S2065: If all first candidate grids (grids with the highest score) in the left rear area have been traversed, the process returns to S2062. Otherwise, a first candidate grid P1 is taken from the left rear area.

[0141] It is determined whether the distance between Pi and Pl satisfies the vehicle size constraint, and whether the absolute value of the difference between the angles PjPiPl and 90 degrees is less than a set threshold. If both conditions are met, Pl is recorded and the process proceeds to S2066. If both conditions are not met, the process returns to S2065.

[0142] S2066: Record the current optimal carrier columnar component combination PiPjPkPl.

[0143] For example, the first candidate grid Pi in the left front area is used as the first target grid Pi, the first candidate grid Pj in the right front area is used as the first target grid Pj, the first candidate grid Pk in the right rear area is used as the first target grid Pk, and the first candidate grid Pl in the left rear area is used as the first target grid Pl.

[0144] In the above process, the vehicle size constraint means that: the distance d between the two symmetrical supporting parts (that is, the actual distance between the two symmetrical supporting parts) can be pre-configured, and the error threshold d' can be pre-configured. On this basis, if the distance is within the range of [d-d', d+d'], it is determined that the distance satisfies the vehicle size constraint; otherwise, if the distance is not within the range of [d-d', d+d'], it is determined that the distance does not satisfy the vehicle size constraint.

[0145] At this point, the introduction of step 206 is completed, and the subsequent step 207 can be continued.

[0146] Step 207: Determine a first pose of the first vehicle in the forklift coordinate system based on the first target grid, and determine a second pose of the second vehicle in the forklift coordinate system based on the second target grid.

[0147] For example, the first position of the first vehicle in the forklift coordinate system can be determined based on the first target grid in the left front area and the first target grid in the right front area. Alternatively, the first position of the first vehicle in the forklift coordinate system can be determined based on the first target grid in the left rear area and the first target grid in the right rear area.

[0148] For example, the second position and posture of the second vehicle in the forklift coordinate system can be determined based on the second target grid in the left front area and the second target grid in the right front area. Alternatively, the second position and posture of the second vehicle in the forklift coordinate system can be determined based on the second target grid in the left rear area and the second target grid in the right rear area.

[0149] In a possible implementation, the following formula (3) may be used to determine the first pose of the first vehicle in the forklift coordinate system, or to determine the second pose of the second vehicle in the forklift coordinate system.

[0150]

[0151] In formula (3), if it is necessary to determine the first pose of the first vehicle in the forklift coordinate system, (pnt0.x, pnt0.y) represents the center point coordinates of the first target grid in the left front area, (pnt1.x, pnt1.y) represents the center point coordinates of the first target grid in the right front area, or (pnt0.x, pnt0.y) represents the center point coordinates of the first target grid in the left rear area, (pnt1.x, pnt1.y) represents the center point coordinates of the first target grid in the right rear area. Posepod.x, Posepod.y, and Posepod.t represent the first pose of the first vehicle in the forklift coordinate system.

[0152] In formula (3), if it is necessary to determine the second pose of the second vehicle in the forklift coordinate system, (pnt0.x, pnt0.y) represents the center point coordinates of the second target grid in the left front area, (pnt1.x, pnt1.y) represents the center point coordinates of the second target grid in the right front area, or (pnt0.x, pnt0.y) represents the center point coordinates of the second target grid in the left rear area, (pnt1.x, pnt1.y) represents the center point coordinates of the second target grid in the right rear area. Posepod.x, Posepod.y, and Posepod.t represent the second pose of the second vehicle in the forklift coordinate system.

[0153] Step 208: Determine a target posture of the forklift based on the first posture and the second posture.

[0154] Exemplarily, the target position of the forklift in the world coordinate system may be determined by the following steps:

[0155] Step 2081: Based on the first pose of the first vehicle in the forklift coordinate system, the first pose may be converted into a first rotation and translation matrix, that is, the rotation and translation matrix of the first vehicle in the forklift coordinate system.

[0156] For example, see formula (1), represents the first pose of the first vehicle in the forklift coordinate system, Represents the first rotation and translation matrix of the first vehicle in the forklift coordinate system.

[0157] For the convenience of distinction, the first position can be recorded as The first rotation and translation matrix can be written as In the above parameters, pose represents the position, T represents the rotation and translation matrix, rbt represents the forklift coordinate system, cur represents the current position (i.e., the stacking preparation position), pod represents the vehicle, and fork represents that the vehicle is located on the forklift, that is, pod_fork represents the vehicle on the forklift, that is, the first vehicle.

[0158] Step 2082: Based on the second posture of the second vehicle in the forklift coordinate system, the second posture may be converted into a second rotation and translation matrix, that is, the rotation and translation matrix of the second vehicle in the forklift coordinate system.

[0159] For example, see formula (1), represents the second pose of the second vehicle in the forklift coordinate system, Represents the second rotation and translation matrix of the second vehicle in the forklift coordinate system.

[0160] For the convenience of distinction, the second posture can be recorded as The second rotation and translation matrix can be written as In the above parameters, pose represents the position, T represents the rotation and translation matrix, rbt represents the forklift coordinate system, cur represents the current position (i.e., the stacking preparation position), pod represents the vehicle, and stor represents that the vehicle is located in the storage location, i.e., pod_stor represents the vehicle on the storage location, i.e., the second vehicle.

[0161] Step 2083: Convert the posture of the forklift in the world coordinate system into a third rotation and translation matrix.

[0162] For example, see formula (1), represents the position of the forklift in the world coordinate system, Represents the third rotation and translation matrix of the forklift in the world coordinate system. The position and posture of the forklift in the world coordinate system is the current position and posture of the forklift, that is, during the movement of the forklift, the position and posture of the forklift in the world coordinate system can be measured, such as using various types of sensors to measure the position and posture of the forklift in the world coordinate system.

[0163] For the convenience of distinction, the position of the forklift in the world coordinate system can be recorded as The third rotation and translation matrix can be written as In the above parameters, pose represents the position, T represents the rotation and translation matrix, rbt represents the forklift coordinate system, cur represents the current position (i.e., the stacking preparation position), and w represents the world coordinate system, that is, it represents the position and rotation and translation matrix of the forklift coordinate system in the world coordinate system.

[0164] Step 2084: Determine the fourth rotation and translation matrix of the first vehicle in the world coordinate system based on the first rotation and translation matrix and the third rotation and translation matrix. For example, the first rotation and translation matrix is ​​the rotation and translation matrix of the first vehicle in the forklift coordinate system, and the third rotation and translation matrix is ​​the third rotation and translation matrix of the forklift in the world coordinate system. Based on the first rotation and translation matrix and the third rotation and translation matrix, the conversion between the first vehicle and the world coordinate system can be realized to obtain the fourth rotation and translation matrix of the first vehicle in the world coordinate system.

[0165] For example, the fourth rotation and translation matrix of the first vehicle in the world coordinate system can be determined by the following formula: In the above formula, represents the fourth rotation and translation matrix, represents the first rotation and translation matrix, Represents the third rotation-translation matrix.

[0166] Step 2085: Determine the fifth rotation and translation matrix of the second vehicle in the world coordinate system based on the second rotation and translation matrix and the third rotation and translation matrix. For example, the second rotation and translation matrix is ​​the rotation and translation matrix of the second vehicle in the forklift coordinate system, and the third rotation and translation matrix is ​​the third rotation and translation matrix of the forklift in the world coordinate system. Based on the second rotation and translation matrix and the third rotation and translation matrix, the conversion between the second vehicle and the world coordinate system can be realized to obtain the fifth rotation and translation matrix of the second vehicle in the world coordinate system.

[0167] For example, the fifth rotation and translation matrix of the second vehicle in the world coordinate system can be determined using the following formula: In the above formula, represents the fifth rotation and translation matrix, represents the second rotation and translation matrix, Represents the third rotation-translation matrix.

[0168] Step 2086: Determine the sixth rotation and translation matrix of the forklift in the first vehicle coordinate system based on the fourth rotation and translation matrix and the third rotation and translation matrix. For example, the fourth rotation and translation matrix is ​​the rotation and translation matrix of the first vehicle in the world coordinate system, and the third rotation and translation matrix is ​​the rotation and translation matrix of the forklift in the world coordinate system. Based on the fourth rotation and translation matrix and the third rotation and translation matrix, the conversion between the forklift and the first vehicle coordinate system can be realized to obtain the sixth rotation and translation matrix of the forklift in the first vehicle coordinate system.

[0169] For example, the sixth rotation and translation matrix of the forklift in the first vehicle coordinate system can be determined by the following formula: represents the sixth rotation and translation matrix, represents the fourth rotation and translation matrix, Represents the third rotation-translation matrix.

[0170] Step 2087, determine the target rotation and translation matrix of the forklift's target posture (i.e., the forklift's target point) in the world coordinate system based on the sixth rotation and translation matrix and the fifth rotation and translation matrix. For example, the sixth rotation and translation matrix is ​​the forklift's rotation and translation matrix in the first vehicle coordinate system, and the fifth rotation and translation matrix is ​​the second vehicle's rotation and translation matrix in the world coordinate system. Based on the sixth rotation and translation matrix and the fifth rotation and translation matrix, the target rotation and translation matrix of the forklift's target posture in the world coordinate system can be obtained.

[0171] For example, the target rotation and translation matrix of the forklift's target posture in the world coordinate system can be determined using the following formula: In the above formula, represents the target rotation and translation matrix, represents the fifth rotation and translation matrix, Represents the sixth rotation and translation matrix.

[0172] Step 2088: convert the target rotation and translation matrix into the target posture of the forklift in the world coordinate system. The target posture includes the target position and target posture of the forklift in the world coordinate system. The target position is used to control the forklift to move to the position, and the target posture is used to control the posture of the forklift at the target position.

[0173] For example, see formula (1), represents the target position of the forklift in the world coordinate system, Represents the target rotation and translation matrix of the forklift in the world coordinate system.

[0174] In a possible implementation, for steps 2081 to 2088, since the above-mentioned postures are all two-dimensional postures, the angle calculation in the posture can be directly obtained by adding the angles. Taking the calculation of the target posture of the forklift in the world coordinate system as an example, it can be seen from formula (4), [0][2] represents the first row and third column elements of the corresponding matrix, and [1][2] represents the second row and third column elements of the corresponding matrix.

[0175]

[0176] At this point, the introduction of step 208 is completed, and the subsequent step 209 can be continued.

[0177] Step 209: Control the forklift to move to the target position based on the target posture (i.e., the target posture of the forklift in the world coordinate system). After the forklift moves to the target position, determine whether there is a stacking deviation between the first carrier and the second carrier. If so, return to step 202 and repeat the above process until there is no stacking deviation between the first carrier and the second carrier. If not, the forklift can be controlled to stack the first carrier on the second carrier, that is, successfully stack the first carrier on the second carrier to complete the carrier stacking process.

[0178] Exemplarily, the following steps may be used to determine whether there is a stacking deviation between the first carrier and the second carrier:

[0179] Step 2091: Acquire a third target point cloud of the symmetrical support component of the first vehicle and a fourth target point cloud of the symmetrical support component of the second vehicle.

[0180] For example, the original point cloud is collected by radar, and the original point cloud is preprocessed to obtain a reference point cloud. A first reference point cloud and a second reference point cloud are selected from the reference point cloud. Semantic segmentation is performed on each position point in the first reference point cloud, and semantic segmentation is performed on each position point in the second reference point cloud. Position points belonging to the symmetrical support components are selected from the first reference point cloud to form a third target point cloud, and position points belonging to the symmetrical support components are selected from the second reference point cloud to form a fourth target point cloud.

[0181] Exemplarily, the above process can refer to step 202 to step 204, which will not be repeated here.

[0182] Exemplarily, whether there is a stacking deviation between the first carrier and the second carrier may be determined based on the third target point cloud and the fourth target point cloud. For example, the following steps may be used to determine whether there is a stacking deviation.

[0183] Step 2092: Select a third target grid from the grid map based on the third target point cloud, and select a fourth target grid from the grid map based on the fourth target point cloud, wherein the third target grid represents the grid corresponding to the symmetrical support component of the first vehicle in the grid map, and the fourth target grid represents the grid corresponding to the symmetrical support component of the second vehicle in the grid map.

[0184] For example, for each area of ​​the grid map, multiple first candidate grids are selected from the area based on the third target point cloud, and a third target grid (corresponding to the first target grid) is selected from the multiple first candidate grids in the area, so that the third target grids in multiple areas meet the constraint conditions.

[0185] For each area of ​​the grid map, multiple second candidate grids are selected from the area based on the fourth target point cloud, and a fourth target grid (corresponding to the second target grid) is selected from the multiple second candidate grids in the area, so that the fourth target grids of the multiple areas meet the constraint conditions.

[0186] Exemplarily, the above process can refer to step 205-step 206, which will not be repeated here.

[0187] Step 2093: determine the third posture (corresponding to the first posture) of the first vehicle in the forklift coordinate system based on the third target grid, and determine the fourth posture (corresponding to the second posture) of the second vehicle in the forklift coordinate system based on the fourth target grid. For this process, refer to step 207.

[0188] Step 2094: If the deviation between the third target grid and the first target grid is less than the threshold, the deviation between the fourth target grid and the second target grid is less than the threshold, the deviation between the third posture and the first posture is less than the threshold, and the deviation between the fourth posture and the second posture is less than the threshold, it is determined that there is no stacking deviation between the first carrier and the second carrier, and the forklift is controlled to stack the first carrier on the second carrier.

[0189] If at least one of the following conditions is not true: the deviation between the third target grid and the first target grid is less than the threshold, the deviation between the fourth target grid and the second target grid is less than the threshold, the deviation between the third posture and the first posture is less than the threshold, and the deviation between the fourth posture and the second posture is less than the threshold, it is determined that there is a stacking deviation between the first carrier and the second carrier, and it is necessary to return to step 202 and repeat the above process.

[0190] Exemplarily, the deviation between the third target grid and the first target grid may be the distance difference of the grid coordinates, and the deviation between the fourth target grid and the second target grid may be the distance difference of the grid coordinates. The two thresholds for the distance difference (i.e., the distance thresholds) may be the same or different. The deviation between the third posture and the first posture may be the posture difference, and the deviation between the fourth posture and the second posture may be the posture difference. The two thresholds for the posture difference may be the same or different.

[0191] It can be seen from the above technical solutions that in the embodiment of the present application, by determining the target posture of the forklift in the world coordinate system, the first carrier on the forklift is completely aligned with the second carrier in the storage position. When the forklift lowers the carrier, the first carrier is aligned with the second carrier, and there is no risk of tipping over. The safety is relatively high, and the carrier stacking can be effectively completed, and the carrier on the forklift can be accurately stacked on the carrier in the storage position. The deep learning point cloud semantic segmentation technology is used to directly search for columnar components (i.e., symmetrical support components) in the point cloud, which can be applied to more types of vehicles. The scoring grid map is used to obtain the possible center points of the symmetrical support components, which can effectively reduce the interference of point cloud clutter compared to the clustering algorithm. Before stacking and lowering, a stacking detection is performed to calculate the new target point of the forklift, improve the safety of the stacking task, and avoid the tipping over of the stacked carriers due to forklift slipping.

[0192] Based on the same application concept as the above method, a carrier stacking device is proposed in the embodiment of the present application. There is a first carrier on the robot and a second carrier on the storage position. The first carrier needs to be stacked onto the second carrier. Figure 5 FIG. 1 is a schematic diagram of the structure of the device, and the device may include:

[0193] An acquisition module 51 is used to acquire a first target point cloud of a symmetrical support component of a first vehicle and a second target point cloud of a symmetrical support component of a second vehicle, wherein the symmetrical support component is a contact component when two vehicles are stacked; a selection module 52 is used to establish a grid map in a robot coordinate system, wherein the grid map includes a plurality of grids; based on the position of the first target point cloud in the grid map, a first target grid corresponding to the symmetrical support component of the first vehicle in the grid map is selected from the plurality of grids; based on the position of the second target point cloud in the grid map, a second target grid corresponding to the symmetrical support component of the second vehicle in the grid map is selected from the plurality of grids; a determination module 53 is used to determine a first pose of the first vehicle in the robot coordinate system based on the first target grid, and to determine a second pose of the second vehicle in the robot coordinate system based on the second target grid; and a target pose of the robot is determined based on the first pose and the second pose; a control module 54 is used to control the robot to stack the first vehicle onto the second vehicle based on the target pose.

[0194] Exemplarily, when the acquisition module 51 acquires the first target point cloud of the symmetrical supporting component of the first carrier and the second target point cloud of the symmetrical supporting component of the second carrier, it is specifically used to: after controlling the robot to move to the configured stacking preparation position and raising the height of the radar to the configured target height, collect the original point cloud through the radar, and the original point cloud includes the point cloud of the first carrier and the point cloud of the second carrier; wherein the radar is deployed on the toothing of the robot and moves up and down with the toothing; or, the radar is deployed on the moving part of the robot and moves up and down with the moving part; based on the position and posture of the radar in the robot coordinate system, determine the rotation and translation matrix from the radar coordinate system to the robot coordinate system; through the rotation and translation The matrix transforms the original point cloud into the robot coordinate system to obtain a reference point cloud in the robot coordinate system; selects a first reference point cloud within the ROI range of the first vehicle from the reference point cloud, and performs semantic segmentation on each position point in the first reference point cloud to obtain semantic information; selects a second reference point cloud within the ROI range of the second vehicle from the reference point cloud, and performs semantic segmentation on each position point in the second reference point cloud to obtain semantic information; wherein the semantic information indicates whether the position point is a symmetrical support component or not; selects position points belonging to the symmetrical support components from the first reference point cloud to form the first target point cloud, and selects position points belonging to the symmetrical support components from the second reference point cloud to form the second target point cloud.

[0195] Exemplarily, the acquisition module 51 is also used to determine the ROI range of the first vehicle based on the rough pose of the first vehicle in the robot coordinate system, the length dimension of the first vehicle, the width dimension of the first vehicle, the target height of the radar, the length error tolerance value, the width error tolerance value and the height error tolerance value before selecting the first reference point cloud within the ROI range of the first vehicle from the reference point cloud; determine the rough pose of the second vehicle in the robot coordinate system according to the pose of the center point of the storage location and the pose of the robot in the world coordinate system; determine the ROI range of the second vehicle based on the rough pose, the length dimension of the second vehicle, the width dimension of the second vehicle, the target height of the radar, the length error tolerance value and the width error tolerance value.

[0196] Exemplarily, when the selection module 52 selects the first target grid corresponding to the symmetrical support component of the first vehicle in the grid map from the multiple grids based on the position of the first target point cloud in the grid map, it is specifically used to: divide the grid map into multiple areas, each area including multiple grids; for each area, determine the first score value of the grid based on the distance between the grid in the area and the position of the first target point cloud in the grid map, select multiple first candidate grids from the area based on the first score value of each grid; and determine the first target grid from the multiple first candidate grids in the area.

[0197] Exemplarily, when the selection module 52 determines the first score value of the grid based on the distance between the grid in the area and the position of the first target point cloud in the grid map, it is specifically used to: determine, for each position point of the first target point cloud, the grid coordinates of the position point in the grid map; for each grid in the area, based on the distance between the grid and the grid coordinates, determine the score value corresponding to the position point in the grid; wherein, the smaller the distance, the larger the score value; and count the first score value of each grid in the area, the first score value being the sum of the score values ​​corresponding to each position point of the first target point cloud in the grid.

[0198] Exemplarily, when determining the first target grid from multiple first candidate grids in the area, each area corresponds to a first target grid, and the first target grids corresponding to the multiple areas satisfy the constraint conditions; the multiple areas include a left front area, a right front area, a left rear area, and a right rear area; the first target grids corresponding to the multiple areas satisfy the constraint conditions, including: the distance between the first target grid Pi in the left front area and the first target grid Pj in the right front area satisfies the vehicle size constraint; the distance between the first target grid Pj and the first target grid Pk in the right rear area satisfies the vehicle size constraint, and the absolute value of the difference between the first angle and 90 degrees is less than a threshold; wherein the first angle is the angle between the first target grid Pi, the first target grid Pj, and the first target grid Pk; the distance between the first target grid Pi and the first target grid Pl in the left rear area satisfies the vehicle size constraint, and the absolute value of the difference between the second angle and 90 degrees is less than a threshold; wherein the second angle is the angle between the first target grid Pj, the first target grid Pi, and the first target grid Pl.

[0199] Exemplarily, when the determination module 53 determines the target posture of the robot based on the first posture and the second posture, it is specifically used to: convert the first posture into a first rotation and translation matrix, convert the second posture into a second rotation and translation matrix, and convert the posture of the robot in the world coordinate system into a third rotation and translation matrix; determine the fourth rotation and translation matrix of the first vehicle in the world coordinate system based on the first rotation and translation matrix and the third rotation and translation matrix; determine the fifth rotation and translation matrix of the second vehicle in the world coordinate system based on the second rotation and translation matrix and the third rotation and translation matrix; determine the sixth rotation and translation matrix of the robot in the first vehicle coordinate system based on the fourth rotation and translation matrix and the third rotation and translation matrix; determine the target rotation and translation matrix of the target posture of the robot in the world coordinate system based on the sixth rotation and translation matrix and the fifth rotation and translation matrix, and convert the target rotation and translation matrix into the target posture of the robot in the world coordinate system.

[0200] Exemplarily, when the control module 54 controls the robot to stack the first carrier onto the second carrier based on the target posture, it is specifically used to: control the robot to move to a target position point based on the target posture; after the robot moves to the target position point, obtain a third target point cloud of the symmetrical supporting parts of the first carrier and a fourth target point cloud of the symmetrical supporting parts of the second carrier; determine whether there is a stacking deviation between the first carrier and the second carrier based on the third target point cloud and the fourth target point cloud; if not, control the robot to stack the first carrier onto the second carrier.

[0201] Exemplarily, when the control module 54 determines whether there is a stacking deviation between the first vehicle and the second vehicle based on the third target point cloud and the fourth target point cloud, it is specifically used to: select a third target grid from the grid map based on the third target point cloud, and select a fourth target grid from the grid map based on the fourth target point cloud; determine a third posture of the first vehicle in the robot coordinate system based on the third target grid, and determine a fourth posture of the second vehicle in the robot coordinate system based on the fourth target grid; if the deviation between the third target grid and the first target grid is less than a threshold, the deviation between the fourth target grid and the second target grid is less than a threshold, the deviation between the third posture and the first posture is less than a threshold, and the deviation between the fourth posture and the second posture is less than a threshold, it is determined that there is no stacking deviation between the first vehicle and the second vehicle.

[0202] Based on the same application concept as the above method, an electronic device is proposed in the embodiment of the present application, see Figure 6As shown, it includes: a processor 61 and a machine-readable storage medium 62, wherein the machine-readable storage medium 62 stores machine-executable instructions that can be executed by the processor 81; the processor 61 is used to execute the machine-executable instructions to implement the carrier stacking method disclosed in the above example of this application.

[0203] Based on the same application concept as the above method, an embodiment of the present application also provides a machine-readable storage medium, on which a number of computer instructions are stored. When the computer instructions are executed by a processor, the carrier stacking method disclosed in the above example of the present application can be implemented.

[0204] The above-mentioned machine-readable storage medium may be any electronic, magnetic, optical or other physical storage device, which may contain or store information, such as executable instructions, data, etc. For example, the machine-readable storage medium may be: RAM (Radom Access Memory), volatile memory, non-volatile memory, flash memory, storage drive (such as hard disk drive), solid state drive, any type of storage disk (such as CD, DVD, etc.), or similar storage medium, or a combination thereof.

[0205] Based on the same application concept as the above method, an embodiment of the present application further provides a carrier stacking system, the system comprising a dispatching device and a robot, the robot being equipped with a radar, the robot being equipped with a first carrier, and the storage being equipped with a second carrier; wherein the radar is deployed on a toothing of the robot and moves up and down with the toothing; or, the radar is deployed on a moving part of the robot and moves up and down with the moving part;

[0206] The scheduling device is used to obtain the configured stacking preparation position and the configured target height, and send the stacking preparation position and the target height to the robot;

[0207] The robot is used to move to the stacking preparation position and raise the height of the radar to the target height; collect original point clouds through the radar, and obtain a first target point cloud of the symmetrical support component of the first carrier and a second target point cloud of the symmetrical support component of the second carrier based on the original point cloud, where the symmetrical support component is a contact component when two carriers are stacked;

[0208] A grid map is established in the robot coordinate system, wherein the grid map includes multiple grids; based on the position of the first target point cloud in the grid map, a first target grid corresponding to the symmetrical support component of the first vehicle in the grid map is selected from the multiple grids; based on the position of the second target point cloud in the grid map, a second target grid corresponding to the symmetrical support component of the second vehicle in the grid map is selected from the multiple grids; a target posture of the robot is determined based on the first target grid and the second target grid; and based on the target posture, the robot is controlled to stack the first vehicle onto the second vehicle.

[0209] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A carrier stacking method, It is characterized in that There is a first carrier on the robot, and there is a second carrier on the storage location, and the method includes: Acquire a first target point cloud of a symmetrical support component of a first carrier and a second target point cloud of a symmetrical support component of a second carrier, wherein the symmetrical support component is a contact component when two carriers are stacked; Establishing a grid map in the robot coordinate system, wherein the grid map includes a plurality of grids; Based on the position of the first target point cloud in the grid map, a first target grid corresponding to the symmetrical support component of the first vehicle in the grid map is selected from the multiple grids; based on the position of the second target point cloud in the grid map, a second target grid corresponding to the symmetrical support component of the second vehicle in the grid map is selected from the multiple grids; Determine a first pose of the first vehicle in a robot coordinate system based on the first target grid, and determine a second pose of the second vehicle in a robot coordinate system based on the second target grid; Determining a target posture of the robot based on the first posture and the second posture; The robot is controlled based on the target posture to stack the first carrier onto the second carrier.

2. The method according to claim 1, It is characterized in that The step of acquiring a first target point cloud of a symmetrical support component of a first vehicle and a second target point cloud of a symmetrical support component of a second vehicle comprises: After controlling the robot to move to the configured stacking preparation position and raising the height of the radar to the configured target height, collecting the original point cloud through the radar, the original point cloud includes the point cloud of the first carrier and the point cloud of the second carrier; wherein the radar is deployed on the toothing of the robot and moves up and down with the toothing; or, the radar is deployed on the moving part of the robot and moves up and down with the moving part; Based on the position and posture of the radar in the robot coordinate system, determine the rotation and translation matrix from the radar coordinate system to the robot coordinate system; transform the original point cloud to the robot coordinate system through the rotation and translation matrix to obtain a reference point cloud in the robot coordinate system; A first reference point cloud within the ROI range of the first vehicle is selected from the reference point cloud, and semantic segmentation is performed on each position point in the first reference point cloud to obtain semantic information; a second reference point cloud within the ROI range of the second vehicle is selected from the reference point cloud, and semantic segmentation is performed on each position point in the second reference point cloud to obtain semantic information; wherein the semantic information indicates whether the position point is a symmetrical support component or not a symmetrical support component; Position points belonging to the symmetrical support components are selected from the first reference point cloud to form a first target point cloud, and position points belonging to the symmetrical support components are selected from the second reference point cloud to form a second target point cloud.

3. The method according to claim 2, It is characterized in that Before selecting the first reference point cloud within the ROI range of the first vehicle from the reference point cloud, the method further includes: Determine the ROI range of the first vehicle based on the rough pose of the first vehicle in the robot coordinate system, the length dimension of the first vehicle, the width dimension of the first vehicle, the target height of the radar, the length error tolerance value, the width error tolerance value, and the height error tolerance value; According to the posture of the center point of the storage location and the posture of the robot in the world coordinate system, the rough posture of the second vehicle in the robot coordinate system is determined; based on the rough posture, the length dimension of the second vehicle, the width dimension of the second vehicle, the target height of the radar, the length error tolerance value and the width error tolerance value, the ROI range of the second vehicle is determined.

4. The method according to claim 1, It is characterized in that The selecting, based on the position of the first target point cloud in the grid map, a first target grid corresponding to the symmetrical support component of the first vehicle in the grid map from the plurality of grids comprises: Dividing the grid map into a plurality of regions, each region including a plurality of grids; For each region, determine a first score value of the grid based on a distance between a grid in the region and a position of the first target point cloud in the grid map, and select a plurality of first candidate grids from the region based on the first score value of each grid; The first target grid is determined from a plurality of first candidate grids in the area.

5. The method according to claim 4, It is characterized in that Determining a first score value of the grid based on a distance between a grid in the area and a position of the first target point cloud in the grid map includes: For each position point of the first target point cloud, determine the grid coordinates of the position point in the grid map; for each grid in the area, determine the score value corresponding to the position point in the grid based on the distance between the grid and the grid coordinates; wherein, the smaller the distance, the larger the score value; A first score value of each grid in the area is counted, where the first score value is the sum of the score values ​​corresponding to each position point of the first target point cloud in the grid.

6. The method according to claim 5, It is characterized in that When determining the first target grid from the plurality of first candidate grids in the region, each region corresponds to a first target grid, and the first target grids corresponding to the plurality of regions satisfy the constraint condition; The multiple areas include a left front area, a right front area, a left rear area, and a right rear area; The first target grids corresponding to the multiple regions satisfy the constraint conditions, including: The distance between the first target grid Pi in the left front area and the first target grid Pj in the right front area satisfies the vehicle size constraint; The distance between the first target grid Pj and the first target grid Pk of the right rear area satisfies the vehicle size constraint, and the absolute value of the difference between the first angle and 90 degrees is less than a threshold; wherein the first angle is the angle between the first target grid Pi, the first target grid Pj, and the first target grid Pk; The distance between the first target grid Pi and the first target grid Pl in the left rear area satisfies the vehicle size constraint, and the absolute value of the difference between the second angle and 90 degrees is less than a threshold; wherein the second angle is the angle between the first target grid Pj, the first target grid Pi, and the first target grid Pl.

7. The method according to claim 1, It is characterized in that The step of determining a target posture of the robot based on the first posture and the second posture comprises: Convert the first posture into a first rotation-translation matrix, convert the second posture into a second rotation-translation matrix, and convert the posture of the robot in the world coordinate system into a third rotation-translation matrix; Determine a fourth rotation and translation matrix of the first vehicle in the world coordinate system based on the first rotation and translation matrix and the third rotation and translation matrix; determine a fifth rotation and translation matrix of the second vehicle in the world coordinate system based on the second rotation and translation matrix and the third rotation and translation matrix; Based on the fourth rotation and translation matrix and the third rotation and translation matrix, the sixth rotation and translation matrix of the robot in the first vehicle coordinate system is determined; based on the sixth rotation and translation matrix and the fifth rotation and translation matrix, the target rotation and translation matrix of the robot's target posture in the world coordinate system is determined, and the target rotation and translation matrix is ​​converted into the robot's target posture in the world coordinate system.

8. The method according to claim 1, It is characterized in that The controlling the robot to stack the first carrier onto the second carrier based on the target posture includes: Controlling the robot to move to a target position point based on the target posture; After the robot moves to the target position, obtaining a third target point cloud of the symmetrical support component of the first carrier and a fourth target point cloud of the symmetrical support component of the second carrier; Determine whether there is a stacking deviation between the first carrier and the second carrier based on the third target point cloud and the fourth target point cloud; If not, controlling the robot to stack the first carrier onto the second carrier; If so, the process returns to execute the operation of acquiring the first target point cloud of the symmetrical support component of the first vehicle and the second target point cloud of the symmetrical support component of the second vehicle.

9. The method according to claim 8, It is characterized in that The determining whether there is a stacking deviation between the first carrier and the second carrier based on the third target point cloud and the fourth target point cloud includes: Based on the position of the third target point cloud in the grid map, a third target grid corresponding to the symmetrical support component of the first vehicle in the grid map is selected from the grid map; based on the position of the fourth target point cloud in the grid map, a fourth target grid corresponding to the symmetrical support component of the second vehicle in the grid map is selected from the grid map; a third pose of the first vehicle in the robot coordinate system is determined based on the third target grid, and a fourth pose of the second vehicle in the robot coordinate system is determined based on the fourth target grid; If the deviation between the third target grid and the first target grid is less than a threshold, the deviation between the fourth target grid and the second target grid is less than a threshold, the deviation between the third posture and the first posture is less than a threshold, and the deviation between the fourth posture and the second posture is less than a threshold, it is determined that there is no stacking deviation between the first vehicle and the second vehicle.

10. A carrier stacking device, It is characterized in that There is a first carrier on the robot, and there is a second carrier on the storage location, and the device includes: An acquisition module, used to acquire a first target point cloud of a symmetrical support component of a first carrier and a second target point cloud of a symmetrical support component of a second carrier, wherein the symmetrical support component is a contact component when two carriers are stacked; A selection module is used to establish a grid map in the robot coordinate system, wherein the grid map includes a plurality of grids; based on the position of the first target point cloud in the grid map, select a first target grid corresponding to the symmetrical support component of the first vehicle in the grid map from the plurality of grids; based on the position of the second target point cloud in the grid map, select a second target grid corresponding to the symmetrical support component of the second vehicle in the grid map from the plurality of grids; A determination module, configured to determine a first pose of the first vehicle in a robot coordinate system based on the first target grid, determine a second pose of the second vehicle in a robot coordinate system based on the second target grid; and determine a target pose of the robot based on the first pose and the second pose; A control module is used to control the robot to stack the first carrier onto the second carrier based on the target posture.

11. An electronic device, It is characterized in that include: a processor and a machine-readable storage medium storing machine-executable instructions executable by the processor; The processor is used to execute machine executable instructions to implement the method described in any one of claims 1-9.

12. A carrier stacking system, It is characterized in that The system includes a dispatching device and a robot, wherein a radar is deployed on the robot, a first carrier is located on the robot, and a second carrier is located on the storage position; wherein the radar is deployed on the inserting gear of the robot and moves up and down with the inserting gear; or, the radar is deployed on a moving part of the robot and moves up and down with the moving part; The scheduling device is used to obtain the configured stacking preparation position and the configured target height, and send the stacking preparation position and the target height to the robot; The robot is used to move to the stacking preparation position and raise the height of the radar to the target height; collect original point clouds through the radar, and obtain a first target point cloud of the symmetrical support component of the first carrier and a second target point cloud of the symmetrical support component of the second carrier based on the original point cloud, where the symmetrical support component is a contact component when two carriers are stacked; A grid map is established in the robot coordinate system, wherein the grid map includes multiple grids; based on the position of the first target point cloud in the grid map, a first target grid corresponding to the symmetrical support component of the first vehicle in the grid map is selected from the multiple grids; based on the position of the second target point cloud in the grid map, a second target grid corresponding to the symmetrical support component of the second vehicle in the grid map is selected from the multiple grids; a target posture of the robot is determined based on the first target grid and the second target grid; and based on the target posture, the robot is controlled to stack the first vehicle onto the second vehicle.

Citation Information

Patent Citations

  • Robot, control method and control device thereof, calibration method and calibration control device, and storage medium

    CN114055444A

  • Raster map updating method and device, electronic equipment and storage medium

    CN114299145A

  • Pose determination method and device of mobile robot

    CN114911232A

  • Unstacking positioning method, device and system and storage medium

    CN116468791A

  • Robot positioning method, robot and readable storage medium

    CN117330083A