A mechanical arm path planning method and system for rule-oriented box unloading

Abstract

The application provides a mechanical arm path planning method and system for rule box unloading, comprising a cargo coordinate processing module and a mechanical arm path planning module, the cargo coordinate processing module can obtain start point coordinates and end point coordinates, and transmit the processed start point coordinates and end point coordinates to the mechanical arm path planning module, and the mechanical arm path planning module can create a moving path of the mechanical arm according to the start point coordinates and the end point coordinates; wherein the cargo coordinate processing module can obtain the posture when the mechanical arm grabs the cargo and the sequence of grabbing the cargo, the application discloses a mechanical arm path planning method for rule box unloading, the path search speed of the application is fast, the result of path search is optimized, complex environment can be quickly modeled, and the application is suitable for path search tasks under different placing conditions of cargo by using different clamps.

Description

Technical Field

[0001] This invention relates to the field of robot control, and in particular to a method and system for path planning of a robotic arm for unloading regular boxes. Background Technology

[0002] In automated factory scenarios, it's common to encounter situations where container trucks are transporting goods in regular-shaped containers. When using robotic arms for unloading, planning the robotic arm's operating path and ensuring the safety of the unloading process are pressing issues. When the robotic arm operates, obstacles such as other hardware, container walls, and other goods may exist along its path and around the target object. Therefore, collisions with these obstacles must be considered during path planning. Furthermore, in this scenario, gripping stability is crucial to prevent the goods from slipping out of the gripper. The robotic arm's end effector must maintain a relatively fixed posture, further complicating path planning and impacting the successful planning of the robot's movement path. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for path planning of a robotic arm for unloading regular containers, which optimizes the path search results of the robotic arm.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a robotic arm path planning method for unloading regular container boxes, comprising the following steps: S1: obtaining the starting point coordinates and the ending point coordinates; S2: using the starting point coordinates as the current coordinates of the robotic arm; S3: using the current coordinates of the robotic arm as the parent node, moving the robotic arm a first distance in a selected direction from the parent node, and using the moved coordinates as the child node; S4: determining whether the robotic arm collides during its movement to the child node; if the robotic arm collides, changing the coordinates of the robotic arm to the coordinates of the parent node and repeating step S3; S5: determining whether the current child node exists within a first range. S6: Determine whether the remaining parent nodes exceed a first preset value. If they do, discard all parent nodes and repeat step S2. S7: Determine whether the current child node is within the operation range of the endpoint coordinates. If it is within the operation range, connect all parent nodes sequentially in chronological order to form the movement path of the robotic arm and confirm the way the robotic arm grabs the goods. If it is not within the operation range of the endpoint coordinates, repeat step S3. S8: Store the movement path of the robotic arm in the storage device.

[0005] Furthermore, the coordinates of the robotic arm are the coordinates of the center point of the robotic arm's flange.

[0006] Furthermore, determining whether a collision occurs during the movement of the robotic arm to the child node includes: acquiring the motion trajectories of the robotic arm and the cargo; acquiring an obstacle model; and determining whether the robotic arm will collide or interfere with the obstacle model during its movement.

[0007] Further, the step of moving the robotic arm from the parent node to the selected direction by a first distance includes: determining whether there is a historical path between the starting point coordinates and the ending point coordinates and whether the current parent node is located within the historical path; if there is a historical path and the current parent node is located within the historical path, then the robotic arm moves from the parent node along the direction of the historical path by a first distance; otherwise, the robotic arm moves from the parent node to a direction randomly selected from the selectable directions by a first distance.

[0008] This invention also discloses a robotic arm path planning system for unloading regular-shaped containers, including a cargo coordinate processing module and a robotic arm path planning module. The cargo coordinate processing module can obtain the starting point coordinates and the ending point coordinates, process the starting point coordinates and the ending point coordinates, and transmit them to the robotic arm path planning module. The robotic arm path planning module can create the movement path of the robotic arm based on the starting point coordinates and the ending point coordinates. The cargo coordinate processing module can obtain the posture of the robotic arm when grasping the cargo and the order in which the cargo is grasped.

[0009] Furthermore, the cargo coordinate processing module includes a gripping posture judgment module, a gripping position conversion module, and a gripping order sorting module. The gripping position conversion module can convert the center point coordinates of the cargo according to the gripping method of the cargo. The gripping posture judgment module can determine the posture of the robotic arm when gripping the cargo. The gripping order sorting module can determine the order in which the cargo is gripped.

[0010] Furthermore, the robotic arm path planning module includes an obstacle judgment module, a gripper selection module, a basic path calculation module, and a special path planning module; the obstacle judgment module can determine whether the robotic arm collides with the obstacle model built into the obstacle judgment module during its movement; the gripper selection module has a built-in gripper model, and the gripper selection module can provide the robotic arm with an abstract model of the gripper; the basic path calculation module is used to calculate the movement path of the robotic arm; the special path planning module can plan the path for the robotic arm to grip the goods.

[0011] Analysis reveals that this invention discloses a robotic arm path planning method and system for unloading regular-shaped containers. The invention features fast path search speed and optimized path search results. In practical use, it allows the robotic arm to complete path planning for the next cargo grabbing point immediately after grabbing the previous cargo, achieving seamless connection. Simultaneously, it exhibits high stability, performing well even in enclosed environments and with complex obstacles. By modeling the environment and obstacles within the system, the system verifies whether each part of the robotic arm will collide with obstacles during the path search phase, ensuring the planned path can be executed smoothly. Through basic models of different types of obstacles, complex environments can be quickly modeled. It is applicable to path search tasks using different grippers and facing different cargo placement scenarios. Attached Figure Description

[0012] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. Wherein:

[0013] Figure 1 A flowchart of a robotic arm path planning method for unloading regular container boxes according to an embodiment of the present invention.

[0014] Figure 2 A structural block diagram of a robotic arm path planning system for unloading regular container boxes according to an embodiment of the present invention. Detailed Implementation

[0015] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. Various examples are provided by way of explanation and not by way of limitation. Indeed, those skilled in the art will recognize that modifications and variations can be made to the invention without departing from its scope or spirit. For example, a feature shown or described as part of one embodiment may be used in another embodiment to produce yet another embodiment. Therefore, it is desirable that the invention encompass such modifications and variations falling within the scope of the appended claims and their equivalents.

[0016] In the description of this invention, the terms "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," and "bottom," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and do not require the invention to be constructed and operated in a specific orientation; therefore, they should not be construed as limitations on the invention. The terms "connected," "linked," and "set up" used in this invention should be interpreted broadly. For example, they can refer to a fixed connection or a detachable connection; a direct connection or an indirect connection through intermediate components; a wired connection, a radio connection, or a wireless communication signal connection. Those skilled in the art can understand the specific meaning of the above terms according to the specific circumstances.

[0017] The accompanying drawings illustrate one or more examples of the invention. The detailed description uses numerals and letters to refer to features in the drawings. Similar or analogous reference numerals in the drawings and description have been used to refer to similar or analogous parts of the invention. As used herein, the terms “first,” “second,” “third,” and “fourth,” etc., are used interchangeably to distinguish one component from another and are not intended to indicate the location or importance of a single component.

[0018] like Figure 1As shown, according to an embodiment of the present invention, a robotic arm path planning method for unloading regular boxes is provided. The robotic arm has a gripper at its end, and the robotic arm grips the goods through the gripper. The method includes the following steps: S1: Obtain the starting point coordinates and the ending point coordinates; S2: Use the starting point coordinates as the current coordinates of the robotic arm; S3: Use the current coordinates of the robotic arm as the parent node, and move the robotic arm a first distance from the parent node in a selected direction. Moving the robotic arm a first distance from the parent node in the selected direction includes: determining whether there is a historical path between the starting point coordinates and the ending point coordinates and whether the current parent node is located within the historical path. If there is a historical path and the current parent node is located within the historical path, then move the robotic arm a first distance from the parent node along the direction of the historical path. Otherwise, move the robotic arm a first distance from the parent node in a randomly selected direction from the available directions and use the moved coordinates as the child node; S4: Determine whether a collision occurs during the movement of the robotic arm to the child node. Determining whether a collision occurs during the movement of the robotic arm to the child node includes: obtaining the motion trajectory of the robotic arm and the goods; obtaining the obstacle model; and determining whether the robotic arm will collide or interfere with the obstacle model during the movement. If the robotic arm collides, the coordinates of the robotic arm are changed to the coordinates of the parent node and step S3 is repeated; S5: Determine whether there are multiple parent nodes within the first range of the current child node. If there are multiple parent nodes, retain the parent node that is farthest from the child node and discard the remaining parent nodes in the first range; S6: Determine whether the remaining parent nodes exceed the first set value. If they exceed the first set value, discard all parent nodes and repeat step S2; S7: Determine whether the current child node is within the operation range of the endpoint coordinates. If it is within the operation range, connect all parent nodes in chronological order to form the movement path of the robotic arm, and confirm the way the robotic arm grasps the goods. The way the robotic arm grasps the goods includes the following: the model of the gripper, the way the gripper grasps the goods, and the path the gripper grasps the goods. If it is not within the operation range of the endpoint coordinates, repeat step S3; Step S8: Store the movement path of the robotic arm in the storage device. This invention practically obtains the original coordinates of goods through an external vision system and inputs them into the overall path planning system in the form of a string. Then, after processing and planning, the path planning system outputs the planned path in the form of an array of coordinate points.

[0019] Furthermore, the coordinates of the robotic arm are the coordinates of the center point of the robotic arm's flange.

[0020] like Figure 1As shown, after obtaining the gripping coordinates, the robotic arm path planning module first determines the current gripper model through the gripper selection module, and converts the gripping point into the center point of the robotic arm flange through the gripper TCP parameters. Simultaneously, it obtains the coordinates of the unloading start point in the gripper model as the starting point position for path planning. At this point, the start and end points of the path planning have been obtained. The start and end point coordinates are then input into the basic path calculation module, which initiates the path planning process. First, it retrieves the corresponding historical data based on the start point. If corresponding historical data exists, it retrieves the historical data for path searching; otherwise, it creates a new historical data file and writes the content of this search into the historical data file. The RRT* module is then used for path planning. Starting with the current position as the parent node, the node moves randomly in one direction by one step (usually 120mm, adjustable depending on the situation) to become a child node. After acquiring this point, the collision detection function in the obstacle detection module checks for collisions at that node. If a collision occurs, the node is discarded and the previous step is repeated. If no collision occurs, it is determined whether there is a better parent node within a specified range. This is determined by the distance cost between path nodes. If a better parent node is found, the node is connected to the new parent node; otherwise, the current parent node is used. After completing these steps, the search is repeated with this node as the new parent node until a new node enters the target point's radius, completing the search task. The search automatically exits when the number of searched nodes exceeds the maximum number to ensure overall system efficiency. A limit is set for the number of search results; if the number of search results is less than this limit, the path search for the current point continues to ensure path stability.

[0021] Once the basic path search is complete, the special path planning module will determine whether path planning needs to be re-executed for special coordinate points based on the current path search results. If so, path planning will be performed for special cases, such as singularity locations. Simultaneously, the module will plan the fixture approach method and the goods placement path based on cargo parameters. Once the planning for one target point in the target point array is complete, the module automatically proceeds to the planning process for the next target point until path planning for all target points in the array is completed, at which point the completed planned path is output.

[0022] like Figure 2 As shown, the present invention also discloses a robotic arm path planning system for unloading regular containers, including a cargo coordinate processing module and a robotic arm path planning module. The cargo coordinate processing module can obtain the starting point coordinates and the ending point coordinates, process the starting point coordinates and the ending point coordinates, and transmit them to the robotic arm path planning module. The robotic arm path planning module can create the movement path of the robotic arm based on the starting point coordinates and the ending point coordinates. The cargo coordinate processing module can obtain the posture of the robotic arm when grasping the cargo and the order in which the cargo is grasped.

[0023] Furthermore, the cargo coordinate processing module includes a gripping posture judgment module, a gripping position conversion module, and a gripping order sorting module. The gripping position conversion module can convert the center point coordinates of the cargo according to the gripping method of the cargo. The gripping posture judgment module can determine the posture of the robotic arm when gripping the cargo. The gripping order sorting module can determine the order in which the cargo is gripped.

[0024] Furthermore, the robotic arm path planning module includes an obstacle judgment module, a gripper selection module, a basic path calculation module, and a special path planning module. The obstacle judgment module can determine whether the robotic arm collides with the obstacle model built into it during movement. The gripper selection module has a built-in gripper model and can provide the robotic arm with an abstract model of the gripper. The basic path calculation module is used to calculate the movement path of the robotic arm. The special path planning module can plan the path for the robotic arm to grip the goods. To achieve the path planning function, this path planning system typically includes two main modules: a goods coordinate processing module and a robotic arm path planning module. After the original coordinate points of the goods are input, they first enter the goods coordinate processing module, which consists of three sub-modules: a gripping posture judgment module, a gripping position conversion module, and a gripping order sorting module. The goods coordinate information string input by the external vision module is processed and first enters the gripping posture judgment module. This module determines the current placement posture of the goods based on the size information and pose quaternion information of the goods, and transforms the goods pose state into the pose state when the gripper grips the goods. The processed coordinate point array is then input into the gripping position conversion module. Since the input cargo coordinates are the center point coordinates, this position cannot be directly gripped. Therefore, this module determines the gripping method based on the cargo's position within the container before performing the gripping coordinate conversion. Gripping methods are divided into side suction and top suction. In side suction, the front end of the gripper is coplanar with the bottom surface of the cargo; in top suction, the bottom end of the gripper is coplanar with the top surface of the cargo. Therefore, in side suction, this module converts the gripping coordinate points into the center position coordinates of the cargo's bottom edge based on the cargo's dimensions and orientation; in top suction, it converts them into the center position coordinates of the top surface of the container. After completing the gripping position conversion, the processed coordinate point array is input into the gripping order sorting module. This module sorts the cargo gripping coordinate points from left to right and from top to bottom according to the cargo's position within the container. After sorting, the cargo coordinate processing module inputs the sorted coordinate point array into the robotic arm path planning module, which then performs path planning for each gripping point according to the sorted order.

[0025] The robotic arm path planning module comprises four sub-modules: obstacle detection, gripper selection, basic path calculation, and special path planning. The obstacle detection and gripper selection modules pre-store obstacle and gripper-related parameters for later use during path planning. The obstacle detection module includes built-in abstract obstacle models, such as walls, spheres, cuboids, and links, facilitating the modeling of real-world scenarios. It also includes a built-in collision function to calculate collisions between the robotic arm and obstacles at different path points, and to determine if the gripper will collide with the robotic arm while grasping goods, ensuring collision-free operation. The gripper selection module includes an abstract gripper model, encompassing collision models, gripper center point coordinates, unloading start point coordinates, and the range of Euler angles for gripper movement. This model allows for the abstract modeling of different grippers within the system. Before path planning, this module selects the appropriate gripper model based on the actual situation and inputs it into the path planning module. The basic path calculation module incorporates the Progressive Optimal Fast Random Search Tree (RRT*) algorithm. Inputting the starting and ending coordinates allows the module to search for the corresponding robotic arm path. The RRT* algorithm module also includes a search parameter adjustment module, which can modify search parameters in real-time based on the search results, such as the target point radius, search step size, maximum number of nodes, and number of search results. During path search, the basic path calculation module calls the collision detection function in the obstacle detection module to determine if a collision occurs. If a collision occurs, the search continues until a collision-free path is found. After these operations are complete, the module outputs the planned robotic arm path as an array of coordinates. The module also includes a historical data storage module for the random search tree, which calls historical data for each search to improve efficiency and updates it after each search to further iterate and optimize the path. The special path planning module addresses situations where there are special points in the path, such as when the target gripping point is near a singularity point on the robotic arm, and plans how the gripper approaches the goods and how to place the goods after returning to the starting point. When the target gripping point is near a singularity, the robotic arm cannot reach the target point directly. In this case, the special path planning module will obtain the points on the path closest to the target point and replan the robotic arm's running path from these points to ensure that the gripper reaches the cargo gripping point in a fixed posture to complete the gripping task. When the gripper approaches the cargo, the module will plan how the gripper should approach the cargo based on the cargo's specific position in the container, such as approaching the cargo directly from above or from a diagonal rear. After the gripper returns to the starting point, the module will plan the specific path for the robotic arm to place the cargo based on the cargo's specific dimensions and its relative position to the conveyor belt.After the cargo coordinate processing module inputs the processed cargo grabbing coordinate point array into the robotic arm path planning module, the robotic arm path planning module will perform path planning for the grabbing coordinate points according to the order in the array.

[0026] Compared to existing technologies, this invention offers faster path search speed. Traditional RRT algorithms suffer from slow search speed and suboptimal results when facing 3D searches. This invention optimizes the path search results. In practical applications, the robotic arm can plan the path to the next cargo grabbing point immediately after completing the grabbing of the previous cargo, achieving seamless transition. It also boasts high stability, performing well even in enclosed environments and with complex obstacles. By modeling the environment and obstacles within the system, the path search phase verifies whether each part of the robotic arm will collide with obstacles, ensuring the planned path can be executed smoothly. Complex environments can be quickly modeled using basic obstacle models of different types. It is suitable for path search tasks using different grippers and facing different cargo placement scenarios.

[0027] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A path planning method for a robotic arm for unloading regular-shaped containers, characterized in that, Includes the following steps: S1: Obtain the coordinates of the starting point and the ending point; S2: Use the starting point coordinates as the current coordinates of the robotic arm; S3: Using the current coordinates of the robotic arm as the parent node, move the robotic arm a first distance from the parent node in the selected direction, and use the coordinates after the movement as the child node; S4: Determine whether the robotic arm collides with the child node during its movement. If the robotic arm collides with the child node, change the coordinates of the robotic arm to the coordinates of the parent node and repeat step S3. S5: Determine whether there are multiple parent nodes within the first range of the current child node. If there are multiple parent nodes, retain the parent node that is farthest from the child node and discard the remaining parent nodes within the first range. S6: Determine whether the remaining parent nodes exceed the first set value. If they exceed the first set value, discard all parent nodes and repeat step S2. S7: Determine whether the current child node is within the operation range of the endpoint coordinates. If it is within the operation range, connect all the parent nodes in chronological order to form the movement path of the robotic arm and confirm the way the robotic arm grabs the goods. If it is not within the operation range of the endpoint coordinates, repeat step S3. The step of moving the robotic arm a first distance from the parent node in the selected direction includes: Determine whether a historical path exists between the starting point coordinates and the ending point coordinates, and whether the current parent node is located within the historical path. If the historical path exists and the current parent node is located within the historical path, then move the robotic arm a first distance from the parent node along the direction of the historical path; otherwise, move the robotic arm a first distance from the parent node by randomly selecting one direction from the selectable directions.

2. The robotic arm path planning method for unloading regular-shaped containers according to claim 1, characterized in that, The coordinates of the robotic arm are the coordinates of the center point of the robotic arm's flange.

3. The robotic arm path planning method for unloading regular-shaped containers according to claim 1, characterized in that, The determination of whether a collision occurs during the movement of the robotic arm to the child node includes: Acquire the motion trajectories of the robotic arm and the cargo; Obtain obstacle models; Determine whether the robotic arm will collide or interfere with the obstacle model during its movement.

4. The robotic arm path planning method for unloading regular-shaped containers according to claim 1, characterized in that, It also includes step S8: storing the movement path of the robotic arm in a storage device.

5. A robotic arm path planning system for unloading regular-shaped containers, executing the method as described in claim 1, characterized in that, It includes a cargo coordinate processing module and a robotic arm path planning module. The cargo coordinate processing module can obtain the starting point coordinates and the ending point coordinates, process the starting point coordinates and the ending point coordinates and transmit them to the robotic arm path planning module. The robotic arm path planning module can create the movement path of the robotic arm based on the starting point coordinates and the ending point coordinates. The cargo coordinate processing module can obtain the posture of the robotic arm when grasping cargo and the order in which the cargo is grasped.

6. The robotic arm path planning system for unloading regular-shaped containers according to claim 5, characterized in that, The cargo coordinate processing module includes a gripping posture judgment module, a gripping position conversion module, and a gripping order sorting module. The gripping position conversion module can convert the center point coordinates of the cargo according to the gripping method of the cargo. The gripping posture judgment module can determine the posture of the robotic arm when gripping the cargo. The gripping order sorting module can determine the order in which the cargo is gripped.

7. A robotic arm path planning system for unloading regular-shaped containers according to claim 6, characterized in that, The robotic arm path planning module includes an obstacle detection module, a gripper selection module, a basic path calculation module, and a special path planning module; The obstacle detection module can determine whether the robotic arm collides with the obstacle model built into the obstacle detection module during its movement; The fixture selection module has a built-in fixture model, which can provide an abstract model of the fixture for the robotic arm. The basic path calculation module is used to calculate the movement path of the robotic arm; The special path planning module can plan the path for the robotic arm to pick up the goods.

Images