Injection molding manipulator posture control method and system based on swarm intelligence
By introducing a swarm intelligence mechanism into the artificial potential field method, the repulsive force and influence distance between manipulators are dynamically adjusted, the attitude control of the multi-manipulator system is optimized, and the problems of manipulator collision and path jitter in traditional methods are solved, realizing efficient and stable multi-manipulator collaboration.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-27
- Publication Date
- 2026-03-24
AI Technical Summary
In multi-robot collaborative environments, traditional artificial potential field methods are difficult to achieve high-precision attitude control and efficient collaboration, which may lead to collisions, path jitter, and deviations in motion trajectories between robots, reducing production efficiency and safety.
By introducing a swarm intelligence mechanism, the collaborative convergence factor and task coupling degree among the robotic arms are calculated, the repulsion coefficient and influence distance are dynamically adjusted, dynamic repulsion and influence range are generated, the movement path of the robotic arms is optimized, and attitude control is performed by combining PID control and least squares curve fitting algorithm.
It improves the posture control accuracy, stability and coordination of robot group operations, avoids collisions, and enhances production efficiency and safety.
Smart Images
Figure CN121716042A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of automation control technology. More specifically, this invention relates to a method and system for posture control of an injection molding robot based on swarm intelligence. Background Technology
[0002] With the continuous development of modern injection molding production technology, the collaborative operation of multiple injection molding robots on the same production line and in the same workspace has become an important trend in high-efficiency production. Traditional single-machine control or posture control methods based on preset trajectories can meet basic operational needs in single-robot operation scenarios. However, in multi-robot collaborative environments, due to the complex spatial interactions, dynamic interference, and uneven task allocation among the robots, these traditional methods often struggle to simultaneously ensure safety, accuracy, and collaborative efficiency. When multiple robots operate at high speeds in a confined space, traditional methods may lead to collisions or excessive avoidance between robots, thereby reducing production efficiency, increasing equipment wear, and posing potential risks to production safety. Therefore, how to achieve high-precision posture control and efficient collaboration in multi-robot systems has become a key technical problem that urgently needs to be solved in the field of intelligent injection molding manufacturing.
[0003] Artificial potential field method is widely used in robot path planning and attitude control due to its simplicity, low computational cost, and ability to directly superimpose the target's attraction and obstacle's repulsion to generate motion driving force. In single-machine or low-density operation scenarios, artificial potential field method can effectively guide the robot to move along a predetermined path and avoid obstacles, achieving basic attitude control and motion planning.
[0004] However, in collaborative environments with multiple robotic arms, the traditional artificial potential field method has significant limitations. First, its repulsive and attractive force parameters are mostly fixed values, lacking the ability to adaptively adjust to the dynamic interactions between robotic arms. This can lead to path jitter, excessive avoidance, or deviations from expected trajectories when robotic arms approach each other, thus affecting overall collaborative efficiency. Second, the artificial potential field method is prone to local minima in complex spatial environments, causing robotic arms to stagnate or fail to reach the target position in high-density work areas. This not only reduces attitude control accuracy but also increases the uncertainty of task execution, resulting in low control precision. Summary of the Invention
[0005] To address the problem of low control precision mentioned in the background art, the present invention provides the following solution.
[0006] This invention provides a swarm intelligence-based posture control method for injection molding robots, comprising: acquiring the Euclidean distance and relative velocity between any two injection molding robot end effectors; calculating the cooperative convergence factor of the two injection molding robots, wherein the cooperative convergence factor is positively correlated with the projection component of the relative velocity between the two injection molding robot end effectors along the connecting line and inversely correlated with the Euclidean distance between the two injection molding robot end effectors; acquiring the task coupling degree between the two injection molding robot end effectors; adjusting the repulsion coefficient and influence distance in the artificial potential field method based on the cooperative convergence factor and the task coupling degree to obtain a dynamic repulsion coefficient and a dynamic influence distance; wherein the dynamic repulsion coefficient is directly proportional to the cooperative convergence factor and inversely proportional to the task coupling degree; and the dynamic influence distance is inversely proportional to the square root of the task coupling degree; and outputting the movement path of each injection molding robot based on the adjusted artificial potential field method to control the posture of the injection molding robot.
[0007] The above technical solution, based on the adjusted artificial potential field method, balances safety, task collaboration, and path smoothness in the generated movement path, enabling the robot to maintain a stable and continuous motion trajectory and accurately adjust its posture during high-speed injection molding operations. This significantly improves task execution efficiency, motion accuracy, and overall operational robustness, while also taking into account the safety and intelligent control effects of multi-robot collaborative operations.
[0008] Furthermore, injection molding robots With injection molding robots Co-approach factors between for: , For injection molding robots With injection molding robots The projected component of the relative velocity between end effectors along the line connecting them. For injection molding robots With injection molding robots Euclidean distance between end effectors To preset hyperparameters, For the natural constant An exponential function with base 0.
[0009] The above technical solution incorporates the relative velocity projection and spatial distance between injection molding robots into the calculation of the cooperative approach factor, enabling the robots to maintain motion coordination while possessing adaptive spatial avoidance capabilities. This allows robots with similar motion speeds and close distances to generate strong cooperative constraints, thereby achieving a balance between smooth approach and dynamic collision avoidance during posture adjustment. This avoids oscillations or collisions caused by simple speed or position control, improving the posture control accuracy and overall operational stability during group operations.
[0010] Furthermore, injection molding robots With injection molding robots Task coupling between for: , and They are injection molding robots With injection molding robots The Euclidean distance between the current position and its mission objective point. For the natural constant An exponential function with base 0.
[0011] The aforementioned technical solution establishes task coupling based on the distance between each robotic arm and its target point, enabling stronger collaboration between robotic arms closer to the target point. This allows for dynamic perception and coordination of task progress during group task execution. It promotes closer cooperation among robotic arms nearing task completion, avoiding unnecessary interference between robotic arms at different task stages and improving the adaptability and execution efficiency of overall task allocation and collaborative paths.
[0012] Furthermore, the dynamic repulsion coefficient for: , Based on the fundamental repulsion coefficient value, For injection molding robots With injection molding robots Co-approach factors between them For injection molding robots With injection molding robots The degree of task coupling between them.
[0013] The aforementioned technical solution achieves an adaptive balance of repulsive forces between robotic arms by dynamically adjusting the repulsive force coefficient in conjunction with the cooperative proximity relationship and the degree of task coupling. When two robotic arms are spatially close but have low task correlation, the repulsive force is enhanced to prevent interference; conversely, when the task correlation is high, the repulsive force is weakened to promote collaborative operation. This effectively improves the collision avoidance flexibility and collaborative coordination of a group of robotic arms in complex working environments, ensuring the safety of attitude control and the overall efficiency of task execution.
[0014] Furthermore, the dynamic influence distance for: , and These are the preset maximum and minimum values that affect the distance, respectively. For injection molding robots With injection molding robots The degree of coupling between tasks.
[0015] The aforementioned technical solution dynamically adjusts the influence distance between robotic arms based on the degree of task coupling. This allows robotic arms with higher task correlation to perform collaborative control within a more compact spatial range, while robotic arms with lower task correlation maintain a larger interaction range, thus achieving an adaptive match between the range of action and task relevance. This effectively balances the tightness of collaboration and spatial safety during group operations, improving space utilization in collaborative work, reducing unnecessary interference risks, and enhancing the intelligence and coordination efficiency of overall control.
[0016] Furthermore, the projection component specifically refers to: acquiring the projection from the injection molding robot. Pointing injection molding robot The unit direction vector will be used by the injection molding robot. With injection molding robots The projection component is obtained by multiplying the relative velocity vector between the end effectors with the unit direction vector.
[0017] Furthermore, it also includes standardizing the cooperative convergence factor.
[0018] Furthermore, a PID controller is used to control the posture of the injection molding robot.
[0019] Furthermore, the preset hyperparameter is 0.1.
[0020] Furthermore, it also includes smoothing the movement path using a least-squares curve fitting algorithm.
[0021] The beneficial effects of this invention are as follows: This invention introduces a swarm intelligence mechanism based on the artificial potential field method, comprehensively incorporating the relative speed, spatial distance, and task-related relationships among robotic arms into the attitude control model. This achieves adaptive coordination of cooperative approach, task coupling, and collision avoidance behaviors. It can dynamically adjust the repulsive force and its range of action according to the robotic arm's motion state and task progress, enabling closely cooperating robotic arms to work efficiently in a smaller space, while robotic arms with independent tasks maintain a safe distance to avoid interference and collisions. This further enhances the stability and path continuity of swarm attitude control, thereby significantly improving the coordination, response flexibility, and overall operational efficiency of injection molding robotic arms in multi-machine collaborative operation scenarios. Attached Figure Description
[0022] Figure 1 This is a flowchart illustrating a swarm intelligence-based posture control method for injection molding robots according to an embodiment of the present invention. Detailed Implementation
[0023] An embodiment of a posture control method for injection molding robots based on swarm intelligence.
[0024] like Figure 1 The flowchart shown is a method for posture control of an injection molding robot based on swarm intelligence according to an embodiment of the present invention, which includes the following steps: S1: Obtain the Euclidean distance and relative speed between any two injection molding robot end effectors.
[0025] In a preferred embodiment, by acquiring the coordinate values of each injection molding robot end effector in a three-dimensional coordinate system in real time, the Euclidean distance between any two end effectors can be accurately calculated. This Euclidean distance can intuitively reflect the relative spatial position relationship between each robot, providing basic data support for subsequent collaborative control and collision avoidance.
[0026] Simultaneously, by acquiring the relative velocity between any two end effectors, the instantaneous velocity components along each direction in three-dimensional space are obtained, thus accurately reflecting the dynamic characteristics of the robot during its movement. Through synchronous monitoring of the Euclidean distance and relative velocity between end effectors, precise perception of the coordinated movement state of the injection molding robot group in the workspace can be achieved, effectively improving the collaborative efficiency and movement safety of the robot in high-speed injection molding production processes and avoiding the risk of collisions between end effectors.
[0027] S2: Calculate the cooperative convergence factor and task coupling degree of any two injection molding robots.
[0028] In a preferred embodiment, the injection molding robot With injection molding robots Co-approach factors between for: , For injection molding robots With injection molding robots The projected component of the relative velocity between end effectors along the line connecting them. For injection molding robots With injection molding robots Euclidean distance between end effectors In this embodiment, the preset hyperparameter is 0.1. Of course, it can also be set according to the actual situation. For the natural constant An exponential function with base 0.5. The projection component specifically refers to: data obtained from the injection molding robot... Pointing injection molding robot The unit direction vector will be used by the injection molding robot. With injection molding robots The projection component is obtained by multiplying the relative velocity vector between the end effectors with the unit direction vector.
[0029] By organically combining the relative velocity between injection molding robots and the spatial distance between end effectors, a convergence factor is constructed to describe the cooperative motion of the robots. The projection of the relative velocity onto the line connecting the two robots reflects the dynamic trend of the robots approaching or moving away, while the Euclidean distance reflects their spatial interval. By combining velocity and distance information through nonlinear mapping and weighting, this cooperative convergence factor can automatically enhance the response when the robots approach each other and reduce the impact when the distance is greater or the relative motion slows down, thereby achieving dynamic cooperative perception and regulation of multiple robot groups. This not only ensures the safe distance and motion coordination of the robots in high-speed injection molding operations, but also improves the overall motion efficiency, enabling precise cooperation and collision avoidance in complex spatial environments, and exhibiting significant real-time performance, robustness, and intelligent control effects.
[0030] It also includes standardizing the co-convergent convergence factor, which may be done using Z-score standardization.
[0031] Injection molding robot With injection molding robots Task coupling between for: , and They are injection molding robots With injection molding robots The Euclidean distance between the current position and its mission objective point. For the natural constant An exponential function with base 0.
[0032] By introducing task coupling degree, the distance between the current position of each injection molding robot and its respective task target point is incorporated into the collaborative evaluation, enabling a quantitative description of the collaborative relationships among multiple robots. Task coupling degree is based on the distance between the current position and the target point, employing an exponential decay method to perform a non-linear mapping of the distance. This ensures that when both robots are close to their respective target tasks, their coupling degree significantly increases, reflecting their high degree of synergy in task execution; conversely, when either robot is far from its target, the coupling degree automatically decreases, thus reducing the impact on collaborative decisions. This system can dynamically reflect the tightness of task execution in a multi-robot system, enabling real-time perception and adjustment of task synergy. This helps optimize the collaborative efficiency of robot groups in complex production scenarios, improves the overall accuracy and intelligence of operations, and effectively avoids resource conflicts or motion interference caused by uneven task distribution.
[0033] S3: Based on the aforementioned cooperative approach factor and task coupling degree, the repulsion coefficient and influence distance in the artificial potential field method are adjusted to obtain the dynamic repulsion coefficient and dynamic influence distance.
[0034] In a preferred embodiment, the dynamic repulsion coefficient for: , Based on the fundamental repulsion coefficient value, For injection molding robots With injection molding robots Co-approach factors between them For injection molding robots With injection molding robots The degree of task coupling between them.
[0035] By constructing a dynamic repulsion coefficient, the motion state and task collaboration relationship between robotic arms are organically combined, enabling intelligent adjustment of potential collision risks and coordinated actions in multi-robotic arm systems. Based on the fundamental repulsion force, the dynamic repulsion coefficient is strengthened or weakened according to the approach trend between robotic arms, and weighted according to the degree of task coupling. This ensures that when robotic arms are close to each other and task coupling is low, the repulsion force is significantly enhanced, effectively avoiding collisions and interference; while when robotic arms are spatially close but task coupling is high, the repulsion force automatically weakens, allowing for reasonable collaborative approach. This achieves adaptive management of the dynamic safety distance and motion coordination of robotic arm groups in high-speed injection molding operations, while improving overall task execution efficiency and operational safety, exhibiting real-time performance, robustness, and intelligent control effects.
[0036] The dynamic influence distance for: , and These are the preset maximum and minimum values that affect the distance, respectively. For injection molding robots With injection molding robots The degree of coupling between tasks.
[0037] By dynamically adjusting the influence distance between robotic arms, the task collaboration relationship is incorporated into the control of the spatial range, achieving adaptive collaborative management of the multi-robotic arm system. Specifically, the influence distance automatically scales based on the degree of coupling between the two robotic arms in task execution. When the task coupling is high, the influence distance tends to a minimum, allowing the robotic arms to move closer to each other while maintaining safety to successfully complete the collaborative task. Conversely, when the task coupling is low, the influence distance approaches a maximum, thereby enhancing the spatial repulsion effect between the robotic arms and effectively preventing potential collisions and interference. While ensuring operational safety, this system achieves intelligent adjustment of the collaborative behavior of the robotic arm group, improving the system's task execution efficiency and overall operational stability and flexibility in complex production environments.
[0038] S4: Output the movement path of each injection molding robot based on the adjusted artificial potential field method to control the posture of the injection molding robot.
[0039] In a preferred embodiment, the force distribution and potential motion constraints of each injection molding robot in the workspace are first calculated in real time based on the adjusted artificial potential field method, and the movement path of each robot is generated accordingly, thereby achieving precise control of the robot's posture.
[0040] To further improve the smoothness of the path and the stability of motion, this embodiment incorporates a least-squares curve fitting algorithm to smooth and connect the movement path generated by the artificial potential field method. This allows the robot to avoid abrupt changes, jitter, or drastic acceleration variations during movement, thereby enhancing operational safety and flexibility. Simultaneously, in the attitude control stage, a PID controller is introduced to perform closed-loop adjustment of the robot's joint angles and end effector posture. This enables the robot to accurately track the smoothed path and dynamically correct for real-time deviations, achieving high-precision attitude control.
[0041] This invention introduces a swarm intelligence mechanism based on the artificial potential field method, incorporating the relative motion relationships and task-relatedness between injection molding manipulators into the attitude control process, thus achieving dynamic adaptive adjustment of repulsive force and influence range. It can flexibly adjust the interaction intensity according to the spatial distance, speed relationship, and task progress between the manipulators, ensuring efficient cooperation among closely cooperating manipulators within a limited space, while task-independent manipulators maintain safe avoidance. Through the coupled calculation of cooperative convergence factors, task coupling degree, and dynamic potential field parameters, this invention effectively improves the motion coordination, collision avoidance safety, and trajectory smoothness of swarm manipulators in multi-task parallel scenarios, thereby achieving stable, efficient, and intelligent attitude control in complex working environments.
[0042] In the description of this specification, "multiple" or "several" means at least two, such as two, three or more, unless otherwise explicitly specified.
[0043] While this specification has shown and described numerous embodiments of the invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in the practice of this invention.
Claims
1. A method for posture control of an injection molding robot based on swarm intelligence, characterized in that, include: Obtain the Euclidean distance and relative speed between any two end effectors of an injection molding robot; Calculate the cooperative convergence factor of any two injection molding robots. The cooperative convergence factor is positively correlated with the projected component of the relative velocity between the end effectors of any two injection molding robots in the direction of their connection, and negatively correlated with the Euclidean distance between the end effectors of any two injection molding robots. Obtain the task coupling degree between any two injection molding robot end effectors; based on the cooperative approach factor and task coupling degree, adjust the repulsion coefficient and influence distance in the artificial potential field method to obtain the dynamic repulsion coefficient and dynamic influence distance; the dynamic repulsion coefficient is directly proportional to the cooperative approach factor and inversely proportional to the task coupling degree; the dynamic influence distance is inversely proportional to the square root of the task coupling degree. The movement path of each injection molding robot is output based on the adjusted artificial potential field method to control the posture of the injection molding robot.
2. The posture control method for an injection molding robot based on swarm intelligence according to claim 1, characterized in that, Injection molding robot With injection molding robots Co-approach factors between for: , For injection molding robots With injection molding robots The projected component of the relative velocity between end effectors along the line connecting them. For injection molding robots With injection molding robots Euclidean distance between end effectors To preset hyperparameters, For the natural constant An exponential function with base 0.
3. The posture control method for an injection molding robot based on swarm intelligence according to claim 1, characterized in that, Injection molding robot With injection molding robots Task coupling between for: , and Injection molding robot With injection molding robots The Euclidean distance between the current position and its mission objective point. For the natural constant An exponential function with base 0.
4. The posture control method for an injection molding robot based on swarm intelligence according to claim 1, characterized in that, The dynamic repulsion coefficient for: , Based on the fundamental repulsion coefficient value, For injection molding robots With injection molding robots Co-approach factors between them For injection molding robots With injection molding robots The degree of task coupling between them.
5. The posture control method for an injection molding robot based on swarm intelligence according to claim 1, characterized in that, The dynamic influence distance for: , and These are the preset maximum and minimum values that affect the distance, respectively. For injection molding robots With injection molding robots The degree of coupling between tasks.
6. The posture control method for an injection molding robot based on swarm intelligence according to claim 1, characterized in that, The projection component specifically refers to: acquiring data from the injection molding robot. Pointing injection molding robot The unit direction vector will be used by the injection molding robot. With injection molding robots The projection component is obtained by multiplying the relative velocity vector between the end effectors with the unit direction vector.
7. The posture control method for an injection molding robot based on swarm intelligence according to claim 1, characterized in that, It also includes standardizing the cooperative convergence factor.
8. The posture control method for an injection molding robot based on swarm intelligence according to claim 1, characterized in that, The posture of the injection molding robot is controlled by a PID controller.
9. The posture control method for an injection molding robot based on swarm intelligence according to claim 2, characterized in that, The preset hyperparameter is 0.
1.
10. The posture control method for an injection molding robot based on swarm intelligence according to claim 1, characterized in that, It also includes using a least-squares curve fitting algorithm to smooth the movement path.