Robot stacking rhythm optimization and parallel processing method

Abstract

The invention discloses a robot stacking rhythm optimization and parallel processing method. Comprising the following steps that initialization robot parameter information is established, information collection is carried out, a central scheduling unit produces stacking poses and action paths, the central scheduling unit carries out dynamic task scheduling and distribution, multiple robots carry out parallel execution and real-time optimization, and the stacking robots execute stacking tasks and work completion reset and exception handling are carried out. And the central scheduling unit adopts a heuristic algorithm to perform data analyst on the structured material data set, obtains a stacking pose planning result, and then generates an optimal motion track of the robot according to the stacking pose planning result. According to the method, through full-process beat optimization, the single-task operation time is effectively shortened, meanwhile, the deviation problem in the operation process is solved through a real-time dynamic adjustment mechanism, the beat stability is ensured, and in addition, through parallel processing, efficient collaborative operation of multiple robot units is achieved, and the efficiency advantage of parallel operation of multiple robots is fully exerted.

Description

Technical Field

[0001] This invention belongs to the field of robot control technology, specifically relating to a method for optimizing robot palletizing cycle time and parallel processing. Background Technology

[0002] With the accelerated development of intelligent warehousing and logistics systems, palletizing robots, as core equipment in logistics and production, play an irreplaceable role in improving operational efficiency, ensuring quality stability, and reducing labor costs. However, in intelligent warehousing and logistics systems, the collaborative scheduling of palletizing robots has long faced the common challenges of path planning conflicts and imbalances in task execution efficiency.

[0003] Existing palletizing robots suffer from several drawbacks. First, in traditional single-arm palletizing operations, product subcontracting, shaping, palletizing grouping, and pack pushing are performed sequentially, resulting in cumulative time cycles for each step and hindering overall palletizing efficiency. Second, traditional algorithms take a long time to generate trajectories in complex spaces and do not adequately consider the smoothness of speed during movement, leading to large positioning errors and prolonged start-up and shutdown times. Furthermore, the lack of a scientific task allocation and scheduling mechanism when multiple robots work collaboratively can easily result in job conflicts or resource idleness, failing to fully leverage the efficiency advantages of parallel operations. Consequently, the entire operation cycle is lengthy, impacting the overall operational efficiency of the warehousing material system.

[0004] Therefore, to address the aforementioned technical issues, it is necessary to provide a method for optimizing the cycle time and parallel processing of robot palletizing.

[0005] The information disclosed in this background section is intended only to enhance the understanding of the overall background of the invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a method for optimizing the cycle time and parallel processing of robot palletizing, which can solve the problems of long cycle time and poor coordination of parallel processing of tasks in existing robot palletizing operations, resulting in low work efficiency.

[0007] To achieve the above objectives, a specific embodiment of the present invention provides the following technical solution: A method for optimizing the cycle time and parallel processing of robot palletizing includes the following steps: S1: Establish initial robot parameter information, collect raw data information of palletized materials in real time through the visual perception unit, and transmit the collected information to the central scheduling unit for structured modeling to generate a set of structured material data and a set of operating environment parameters. S2: The central scheduling unit uses a heuristic algorithm to perform data analysis on the structured material data set and obtains the palletizing pose planning results. Then, it generates the optimal motion trajectory of the robot based on the palletizing pose planning results. S3: The central scheduling unit allocates tasks to the robot using a load balancing algorithm based on the palletizing pose planning results, the optimal motion trajectory, and the set of working environment parameters, and generates an outbound sorting queue based on the staggered merging rule. S4: Each robot unit performs palletizing operations in parallel according to the instructions of the central scheduling unit. At the same time, the visual perception unit collects the robot palletizing information in real time, and the central scheduling unit compares the real-time information with the preset operation parameters, corrects the motion trajectory parameters or adjusts the task allocation scheme in real time to optimize the cycle time. S5: After the palletizing task is completed, the central scheduling unit sends a work completion signal, and each robot unit returns to the safe waiting position and performs abnormal handling and recovery.

[0008] In one embodiment of the present invention, the initialization of palletizing robot parameter information in step S1 includes: establishing a three-dimensional workspace model, determining the working range and safety zone of the palletizing robot, and the size information of the goods that the palletizing robot can transport, its moving speed, acceleration, and joint limits, etc. The working range and safety zone of the palletizing robot are established, and the movement path of the palletizing robot is planned within the safety zone. Based on the size and load limit of different palletizing robots, the size of the goods that can be grasped and the weight of the goods that can be loaded are determined for each palletizing robot. Based on the weight and size of the goods loaded by the palletizing robot, the corresponding moving speed and acceleration are configured, etc.

[0009] In one embodiment of the present invention, the original data information of the palletizing material in step S1 includes: material size, weight, posture, conveying speed, conveying distance, and position coordinates. Based on the material's size and weight, a suitable palletizing robot and a material storage area are rationally allocated. The optimal gripping position for the palletizing robot is determined based on the material's posture. The nearest palletizing robot is rationally allocated based on the conveying speed, conveying distance, and position coordinates.

[0010] In one embodiment of the present invention, the palletizing pose planning result in step S2 includes: the palletizing robot's grasping pose, candidate placement pose, and layer identifier. The optimal motion trajectory is dynamically planned using an improved RRT algorithm combined with S-curve acceleration / deceleration control to generate the optimal motion trajectory between each palletizing node. The optimal motion trajectory satisfies the collision-free constraint and has the shortest movement time. Based on the palletizing pose planning result, the palletizing robot grasps the material and moves according to the optimal motion trajectory, placing the material on the candidate placement according to the placement pose. It also promptly generates the layer identifier for the material's storage and transmits it to the central scheduling unit, so that the material information can be retrieved later.

[0011] Among them, the improved RRT algorithm optimizes the random tree expansion direction by introducing a distance heuristic function, shortening the trajectory planning time to the millisecond level; the S-curve acceleration and deceleration control reduces vibration and positioning errors during robot motion and improves motion smoothness by adjusting the rate of acceleration change.

[0012] In one embodiment of the present invention, the task allocation in step S3 includes assigning palletizing tasks of different levels and different areas to corresponding palletizing robot units. The outbound sorting queue includes: clarifying the operation sequence, grasping timing and stacking position of each robot unit; and setting a cycle synchronization threshold to ensure the coordination of the operation cycle of each robot unit and avoid operation conflicts.

[0013] The load balancing algorithm allocates tasks based on the real-time load rate of each robot unit and the distance between the current working position and the target stacking position, ensuring that the load difference of each robot unit does not exceed the preset threshold. The cycle synchronization threshold is dynamically set according to the material conveying speed and the robot movement speed, with a value range of 0.2-0.5 seconds.

[0014] In one embodiment of the present invention, the visual perception unit in step S4 collects real-time robot palletizing information including: robot end-effector posture, material gripping status, and pallet forming status. The central scheduling unit generates real-time feedback data based on the robot end-effector posture, material gripping status, and pallet forming status, and compares the real-time feedback data with preset operating parameters.

[0015] During use, each robot unit performs palletizing operations according to the scheduled object. The vision perception unit collects the robot end posture, material grasping status, and pallet forming status in real time and generates real-time feedback data. When a robot unit is detected to have a grasping deviation, the central scheduling unit corrects the robot end motion trajectory through vision guidance. When a robot unit experiences a delay in operation due to material jamming, the material palletizing task at its level is dynamically allocated to other robot units with lighter loads to ensure that the overall cycle efficiency is not affected.

[0016] In one embodiment of the present invention, the anomaly handling and recovery in step S5 includes: sensor data anomaly detection and robot performance degradation detection, and degrading or shutting down the robot based on the detection data. The robot performance detection structure is divided into minor faults, moderate faults, and severe faults. For minor faults, the robot continues to operate with adaptive parameter adjustments; for moderate faults, the robot undergoes task reassignment and degraded operation; and for severe faults, the robot is safely shut down and requires manual intervention.

[0017] A robot palletizing cycle optimization and parallel processing system is provided to implement a robot palletizing cycle optimization and parallel processing method. The processing system includes: a multi-robot unit, a vision perception unit, a conveyor line unit, a central scheduling unit, and a storage unit.

[0018] In one embodiment of the present invention, the central scheduling unit is electrically connected to the multi-robot unit, the vision perception unit, the conveyor line unit and the storage unit respectively. The storage unit stores a computer program. When the central scheduling unit executes the computer program, it implements the above-mentioned robot palletizing cycle optimization and parallel processing method.

[0019] The multiple robot units are used to receive instructions from the central scheduling unit and control each robot to perform palletizing tasks independently. The vision perception unit is mainly used to collect material information, robot movement path information, and cargo storage location information. The vision perception unit transmits the collected information to the central scheduling unit, generates real-time feedback data on the collected information, and produces response instruction signals to each robot unit.

[0020] Compared with existing technologies, the robot palletizing cycle optimization and parallel processing method of the present invention effectively shortens the single task operation time through full-process cycle optimization, and at the same time, it addresses deviation problems in the operation process through a real-time dynamic adjustment mechanism to ensure cycle stability. In addition, through parallel processing, it realizes efficient collaborative operation of multiple robot units and fully leverages the efficiency advantages of multi-robot parallel operation. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart of a robot palletizing cycle optimization and parallel processing method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of a palletizing robot in one embodiment of the present invention. Detailed Implementation

[0023] To enable those skilled in the art to better understand the technical solutions in this disclosure, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. Based on the embodiments in this disclosure, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this disclosure.

[0024] like Figures 1 to 2 As shown, a robot palletizing cycle optimization and parallel processing method according to an embodiment of the present invention includes the following steps: S1: Establish initial robot parameter information, collect raw data information of palletized materials in real time through the visual perception unit, and transmit the collected information to the central scheduling unit for structured modeling to generate a set of structured material data and a set of operating environment parameters. S2: The central scheduling unit uses a heuristic algorithm to perform data analysis on the structured material data set and obtains the palletizing pose planning results. Then, it generates the optimal motion trajectory of the robot based on the palletizing pose planning results. S3: The central scheduling unit allocates tasks to the robot using a load balancing algorithm based on the palletizing pose planning results, the optimal motion trajectory, and the set of working environment parameters, and generates an outbound sorting queue based on the staggered merging rule. S4: Each robot unit performs palletizing operations in parallel according to the instructions of the central scheduling unit. At the same time, the visual perception unit collects the robot palletizing information in real time, and the central scheduling unit compares the real-time information with the preset operation parameters, corrects the motion trajectory parameters or adjusts the task allocation scheme in real time to optimize the cycle time. S5: After the palletizing task is completed, the central scheduling unit sends a work completion signal, and each robot unit returns to the safe waiting position and performs abnormal handling and recovery.

[0025] Preferred method: In step S1, a multi-view 3D camera is used based on the visual perception unit to achieve accurate acquisition of material information through multi-sensor fusion technology; in the structured modeling process, the material information is associated and mapped with the working environment parameters to generate structured data entries containing material ID, target placement position, and priority.

[0026] like Figures 1 to 2As shown, step S1, based on the initialized palletizing robot parameter information, includes: establishing a 3D workspace model, determining the working range and safety zone of the palletizing robot, as well as the size information of the goods that the palletizing robot can transport, its moving speed, acceleration, and joint limits, etc. The working range and safety zone of the palletizing robot are established, and the movement path of the palletizing robot is planned within the safety zone. Based on the size and load limit of different palletizing robots, the size of the goods that can be grasped and the weight of the goods that can be loaded are determined for each palletizing robot. Based on the weight and size of the goods loaded by the palletizing robot, the corresponding moving speed and acceleration are configured, etc.

[0027] like Figures 1 to 2 As shown, step S1 uses the original data information of the palletized material, including: material size, weight, posture, conveying speed, conveying distance, and position coordinates. Based on the material's size and weight, a suitable palletizing robot is allocated, along with the material's storage area. The optimal gripping position for the palletizing robot is determined based on the material's posture. Finally, the nearest palletizing robot is allocated based on the conveying speed, conveying distance, and position coordinates.

[0028] The material size information collected includes the material's length, width, and height, posture information (e.g., tilt angle ≤ 8°), and position coordinate information (based on the robot's base coordinate system); the encoder and proximity switch of the conveyor line unit collect the material conveying speed information (e.g., 0.3 m / s), conveying spacing (e.g., spacing 400 mm), and pallet position information (e.g., palletizing area coordinates: X=1500 mm, Y=600 mm, Z=900 mm).

[0029] like Figures 1 to 2 As shown, step S2, based on the palletizing pose planning results, includes: the palletizing robot's grasping pose, candidate placement poses, and layer identifiers. Dynamic path planning is then performed using an improved RRT algorithm combined with S-curve acceleration / deceleration control based on the optimal motion trajectory. This generates the optimal motion trajectory between each palletizing node, ensuring that the optimal motion trajectory satisfies collision-free constraints and minimizes movement time. The palletizing robot, based on the palletizing pose planning results, grasps the material and moves it according to the optimal motion trajectory. It places the material onto the candidate placement position according to the placement pose and promptly generates and transmits the layer identifier of the material to the central scheduling unit. This material information is then retrieved later when the material is needed.

[0030] The generated candidate placement poses (e.g., stacked palletizing, 8 pieces per layer, 6 layers in total) and layer identifiers (e.g., layers 1-8) are used. S-curve acceleration / deceleration control is employed to adjust the rate of acceleration change, reducing vibration and positioning errors during robot movement and compressing trajectory planning time to 0.5 seconds per attempt. Preferred: The improved RRT algorithm optimizes the random tree expansion direction by introducing a distance heuristic function, shortening the trajectory planning time to the millisecond level; S-curve acceleration and deceleration control reduces vibration and positioning errors during robot motion and improves motion smoothness by adjusting the rate of acceleration change.

[0031] like Figures 1 to 2 As shown, step S3 includes allocating palletizing tasks of different levels and regions to the corresponding palletizing robot units according to the task allocation, and according to the outbound sorting queue, including: clarifying the operation sequence, grasping timing and stacking position of each robot unit; and setting a cycle synchronization threshold to ensure the coordination of the operation cycle of each robot unit and avoid operation conflicts.

[0032] When using a load balancing algorithm for task allocation, for example, palletizing tasks at levels 1-2 are assigned to robot unit 1, palletizing tasks at levels 3-4 are assigned to robot unit 2, and palletizing tasks at level 5 are assigned to robot unit 3. An outbound sorting queue is generated based on the staggered merging rule and mapped to a scheduling object. The operation sequence of each robot unit is defined (e.g., robot unit 1 starts first, and robot units 2 and 3 start in sequence after an interval of 0.5 seconds), the grasping timing (e.g., the grasping interval is set to 1.8 seconds based on the material conveying distance), and the stacking position. The cycle synchronization threshold is set to 0.4 seconds to ensure the coordination of the operation cycle of each robot unit.

[0033] Preferred method: Based on the load balancing algorithm, tasks are allocated according to the real-time load rate of each robot unit and the distance between the current working position and the target stacking position to ensure that the load difference of each robot unit does not exceed the preset threshold; the cycle synchronization threshold is dynamically set according to the material conveying speed and the robot movement speed, with a value range of 0.2-0.5 seconds.

[0034] like Figures 1 to 2 As shown, in step S4, the robot palletizing information is collected in real time by the visual perception unit, including: robot end-effector posture, material gripping status, and pallet forming status. The central scheduling unit generates real-time feedback data based on the robot end-effector posture, material gripping status, and pallet forming status, and compares the real-time feedback data with the preset operation parameters.

[0035] In use, each robot unit performs palletizing operations according to the scheduled objects. The vision perception unit collects the robot end posture, material grasping status and pallet forming status in real time and generates real-time feedback data. When the grasping deviation of robot unit 2 is detected to be 0.8mm (exceeding the ±0.5mm threshold), the central scheduling unit corrects the robot end motion trajectory through vision guidance. When robot unit 1 experiences a 0.4-second operation delay due to material jamming, the palletizing task of 2 materials at its level 2 is dynamically allocated to robot unit 3 with a lighter load to ensure that the overall cycle efficiency is not affected.

[0036] Preferred: Real-time correction of motion trajectory parameters or adjustment of task allocation includes the following two adjustment strategies: When the material grasping deviation exceeds ±0.5mm, the robot end motion trajectory is corrected through visual guidance; When a robot unit experiences a work delay, some of its low-priority tasks are dynamically allocated to robot units with lighter loads to ensure that the overall cycle efficiency is not affected.

[0037] like Figures 1 to 2 As shown, step S5, based on anomaly handling and recovery, includes: sensor data anomaly detection and robot performance degradation detection. Based on the detected data, the robot is either degraded or shut down. The robot performance detection structure is divided into minor faults, moderate faults, and severe faults. For minor faults, the robot undergoes adaptive parameter adjustment and continues operation; for moderate faults, the robot undergoes task reassignment and degraded operation; and for severe faults, the robot is safely shut down and requires manual intervention.

[0038] A robot palletizing cycle time optimization and parallel processing system is provided to implement a robot palletizing cycle time optimization and parallel processing method. The system includes a multi-robot unit, a vision perception unit, a conveyor line unit, a central scheduling unit, and a storage unit. The central scheduling unit is electrically connected to the multi-robot unit, the vision perception unit, the conveyor line unit, and the storage unit. The storage unit stores a computer program, which is executed by the central scheduling unit to implement the aforementioned robot palletizing cycle time optimization and parallel processing method.

[0039] The multiple robot units are used to receive instructions from the central scheduling unit and control each robot to perform palletizing tasks independently. The vision perception unit is mainly used to collect material information, robot movement path information, and cargo storage location information. The vision perception unit transmits the collected information to the central scheduling unit, generates real-time feedback data on the collected information, and produces response instruction signals to each robot unit.

[0040] It will be apparent to those skilled in the art that this disclosure is not limited to the details of the exemplary embodiments described above, and that this disclosure can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of this disclosure is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within this disclosure. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0041] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for robot palletizing beat optimization and parallel processing, the method comprising: The method comprises the following steps: S1: Establishing initialization robot parameter information, and collecting original data information of the stacking material through a visual perception unit in real time, and transmitting the collected information to a central scheduling unit for structured modeling, generating a structured material data set and a job environment parameter set; S2: The central scheduling unit uses a heuristic algorithm to analyze the structured material data set, and obtains a stacking pose planning result, and then generates an optimal motion trajectory of the robot according to the stacking pose planning result; S3: The central scheduling unit uses a load balancing algorithm to assign tasks to the robot according to the stacking pose planning result, the optimal motion trajectory and the job environment parameter set, and generates an outbound sorting queue based on the interleaving and merging rule; S4: Each robot unit executes the stacking operation in parallel according to the instructions of the central scheduling unit, and collects robot stacking information in real time through the visual perception unit, and compares the real-time information with the preset job parameters by the central scheduling unit to realize beat optimization by correcting the motion trajectory parameters or adjusting the task allocation scheme in real time; S5: After the completion of the stacking task, the central scheduling unit sends a job completion signal, and each robot unit returns to a safe waiting position and performs abnormal processing and recovery.

2. The method of claim 1, wherein, The initialization stacking robot parameter information in the S1 step comprises: establishing a three-dimensional working space model, determining the stacking robot working range and safety area, and the size information, moving speed, acceleration and joint limit of the stacking robot transportable goods.

3. The method of claim 1, wherein, The original data information of the stacking material in the S1 step comprises: material size, weight, attitude, conveying speed, conveying spacing and position coordinates.

4. The method of claim 1, wherein, The stacking pose planning result in the S2 step comprises: stacking robot grasping pose, candidate placement pose and hierarchical identification, the optimal motion trajectory uses an improved RRT algorithm combined with S-curve acceleration and deceleration control for dynamic path planning to generate the optimal motion trajectory between each stacking node, and the optimal motion trajectory satisfies the non-collision constraint and the shortest motion time.

5. The method of claim 1, wherein, The task allocation in the S3 step comprises: allocating the stacking tasks of different levels and different areas to the corresponding stacking robot units, and the outbound sorting queue comprises: the working order, grasping opportunity and stacking position of each robot unit; at the same time, a beat synchronization threshold is set to ensure the cooperation of the working beats of each robot unit and avoid job conflicts.

6. The method of claim 1, wherein, The visual perception unit collects the robot stacking information in real time in the S4 step, which comprises: the robot end attitude, material grasping state and pallet forming state.

7. The method of claim 6, wherein, The central scheduling unit generates real-time feedback data according to the robot end attitude, material grasping state and pallet forming state, and compares the real-time feedback data with the preset job parameters.

8. The method of claim 1, wherein, The abnormal processing and recovery in the S5 step comprises: sensor data anomaly detection and robot performance degradation detection, and the robot is degraded or stopped according to the detection data information.

9. A robot palletizing beat optimization and parallel processing system for implementing a robot palletizing beat optimization and parallel processing method according to any one of claims 1-8, characterized in that, The processing system comprises: a multi-robot unit, a visual perception unit, a conveying line unit, a central scheduling unit and a storage unit.

10. The robotic palletizing beat optimization and parallel processing system of claim 9, wherein, The central scheduling unit is electrically connected with the multi-robot unit, the visual perception unit, the conveying line unit and the storage unit respectively, the storage unit stores a computer program, and the central scheduling unit implements the above-mentioned robot stacking beat optimization and parallel processing method when executing the computer program.

Images