Search

CN-120628092-B - Dynamic path planning learning method and system for industrial robot

CN120628092BCN 120628092 BCN120628092 BCN 120628092BCN-120628092-B

Abstract

S100, defining a working area of an industrial robot, selecting deployment positions of a plurality of identification lamps in each industrial robot, setting colors of the identification lamps to generate combined codes, wherein the combined codes of the industrial robots are different, acquiring video monitoring data comprising all the industrial robots by using video monitoring equipment which is deployed in the working area in advance, intercepting a plurality of snapshots according to a preset step length, and positioning the positions of the industrial robots by the combined codes; and 200, carrying out grid division on the working area, and selecting a reference point. According to the invention, by defining the conflict line, the incremental programming of the industrial robot can be realized, the moving route can be dynamically adjusted, the change of the obstacle can be quickly adapted, and the obstacle avoidance efficiency of the industrial robot is greatly improved.

Inventors

  • LI PENG
  • FU QIANG

Assignees

  • 上海网钜信息科技有限公司

Dates

Publication Date
20260508
Application Date
20250527

Claims (10)

  1. 1. The industrial robot dynamic path planning learning method is characterized by comprising the following steps: S100, defining working areas of industrial robots, selecting deployment positions of a plurality of identification lamps in each industrial robot, setting colors of each identification lamp, generating combined codes, wherein the combined codes of each industrial robot are different, acquiring video monitoring data comprising all industrial robots by using video monitoring equipment which is deployed in the working areas in advance, intercepting a plurality of snapshots according to preset step sizes, and positioning the positions of the industrial robots by the combined codes; S200, carrying out grid division on the working area, selecting a reference point, constructing a coordinate system, receiving a production task uploaded by a preset terminal, finding out a target robot for executing the production task, positioning initial coordinates of the target robot, traversing end point coordinates from the production task, and generating a moving route, wherein the moving route consists of a plurality of grid lines with the same length, at least obtaining a first grid line and a second grid line, and marking out line coordinates of each grid line; And S300, acquiring configuration data of the target robot, calculating expected arrival time of a second grid line based on the position and the length of the grid line when the target robot arrives at the first grid line, generating a lookup table, wherein the lookup table consists of line coordinate items and expected arrival time items of the second grid line, establishing a corresponding relation between the target robot and the lookup table, judging whether the same items exist in all the lookup tables, if so, defining the second grid line corresponding to the same items as a conflict line, searching out an alternative route of the conflict line by utilizing a nearest neighbor algorithm, and updating the moving route.
  2. 2. The method for learning dynamic path planning for industrial robots according to claim 1, wherein the step of defining a working area of the industrial robot, selecting a deployment location of a plurality of identification lamps in each industrial robot, and setting a color of each identification lamp comprises: Formulating a numbering rule, and determining the number of each industrial robot; and integrating the numbers and the combined codes to generate a comparison table.
  3. 3. The method of claim 2, wherein the step of generating a combined code comprises: the flicker frequency of each identification lamp is configured, the flicker time is recorded, and the position is dynamically updated; the production tasks are clustered into several types and the combined codes are adjusted.
  4. 4. The method for learning dynamic path planning of industrial robot according to claim 1, wherein the step of meshing the working area, selecting a reference point, constructing a coordinate system, and receiving a production task uploaded by a preset terminal comprises: setting the priority of each production task, wherein the priority at least comprises high, medium and low; identifying the priority corresponding to each identical item, sorting the identical items according to the order of the priority from high to low, and defining the first identical item of the sorting as a priority item; And opening the use authority of the conflict line to the priority item.
  5. 5. The method according to claim 4, wherein the step of locating the initial coordinates of the target robot and traversing the final coordinates from the production task to generate the moving route comprises: defining two endpoints of the grid line as grid points; based on the configuration data and the production task, an estimated time for the target robot to reach each grid point in the moving route is estimated, and the estimated time is corrected based on the position.
  6. 6. The method of claim 5, wherein defining the second grid line corresponding to the same item as the collision line comprises: Defining two ends of the conflict line as a first node and a second node; And connecting the starting point and the end point by using the grid point and a nearest neighbor algorithm by using the first node as the starting point and the second node as the end point to obtain an alternative route.
  7. 7. The industrial robot dynamic path planning learning method of claim 1, further comprising: collecting sensing data by using sensing equipment which is deployed in a working area in advance, wherein the sensing equipment at least comprises an RFID receiver and a laser radar; and judging whether an obstacle object exists in a preset range of the target robot or not based on the sensing data and the configuration data, and if so, triggering a pre-built avoidance mechanism.
  8. 8. An industrial robot dynamic path planning learning system, the system comprising: The positioning module is used for dividing a working area of the industrial robot, selecting deployment positions of a plurality of identification lamps in each industrial robot, setting colors of each identification lamp, generating combined codes, wherein the combined codes of each industrial robot are different, acquiring video monitoring data comprising all industrial robots by using video monitoring equipment which is deployed in the working area in advance, intercepting a plurality of snapshots according to a preset step length, and positioning the positions of the industrial robots by the combined codes; The calibration module is used for carrying out grid division on the working area, selecting a reference point, constructing a coordinate system, receiving a production task uploaded by a preset terminal, finding out a target robot for executing the production task, positioning the initial coordinate of the target robot, traversing the terminal coordinate from the production task, and generating a moving route, wherein the moving route consists of a plurality of grid lines with the same length, at least obtaining a first grid line and a second grid line, and calibrating the line coordinate of each grid line; And the updating module is used for acquiring configuration data of the target robot, calculating the expected arrival time of the second grid line based on the position and the length of the grid line when the target robot reaches the first grid line, generating a lookup table, wherein the lookup table consists of line coordinate items and expected arrival time items of the second grid line, establishing a corresponding relation between the target robot and the lookup table, judging whether the same items exist in all the lookup tables, if so, defining the second grid line corresponding to the same items as a conflict line, searching out a replacement route of the conflict line by utilizing a nearest neighbor algorithm, and updating the moving route.
  9. 9. The industrial robot dynamic path planning learning system of claim 8, wherein the positioning module comprises: The formulating unit is used for formulating a numbering rule and determining the number of each industrial robot; The generating unit is used for integrating the numbers and the combined codes to generate a comparison table; The recording unit is used for configuring the flicker frequency of each identification lamp, recording the flicker time and dynamically updating the position; and the adjusting unit is used for clustering the production tasks into a plurality of types and adjusting the combined codes.
  10. 10. The industrial robot dynamic path planning learning system of claim 8, wherein the calibration module comprises: a setting unit configured to set a priority of each production task, wherein the priority includes at least high, medium, and low; The identification unit is used for identifying the priority corresponding to each identical item, sequencing the identical items according to the order of the priority from high to low, and defining the identical item in the first sequencing as a priority item; an opening unit, configured to open a usage right of the conflict line to the priority item; An initial defining unit for defining two end points of the grid line as grid points; And a correction unit configured to estimate an estimated time for the target robot to reach each grid point in the moving route based on the configuration data and the production task, and correct the estimated time based on the position.

Description

Dynamic path planning learning method and system for industrial robot Technical Field The invention relates to the technical field of dynamic path planning, in particular to a learning method and a learning system for dynamic path planning of an industrial robot. Background In industrial production, industrial robots, such as logistics transfer robots, material distribution robots, automated guided transport robots, etc., are often used. In the process of executing production tasks, positioning is generally performed through two-dimensional code landmarks, laser radars, visual cameras and the like, but the positioning modes are from individuals, global conditions cannot be observed in real time, and when industrial robots are close to each other in the moving process, avoidance is performed in a stopping mode, so that congestion and collision are easily caused by the avoidance method, and the working efficiency of the industrial robots is greatly influenced. Therefore, "how to use the identification lamp to globally sense the industrial robot" is a technical problem to be solved by the invention. Disclosure of Invention The invention aims to provide a learning method and a learning system for dynamic path planning of an industrial robot, which are used for solving the problem of how to use an identification lamp to perform global perception on the industrial robot in the background technology. In order to achieve the above purpose, the present invention provides the following technical solutions: an industrial robot dynamic path planning learning method, the method comprising: S100, defining working areas of industrial robots, selecting deployment positions of a plurality of identification lamps in each industrial robot, setting colors of each identification lamp, generating combined codes, wherein the combined codes of each industrial robot are different, acquiring video monitoring data comprising all industrial robots by using video monitoring equipment which is deployed in the working areas in advance, intercepting a plurality of snapshots according to preset step sizes, and positioning the positions of the industrial robots by the combined codes; S200, carrying out grid division on the working area, selecting a reference point, constructing a coordinate system, receiving a production task uploaded by a preset terminal, finding out a target robot for executing the production task, positioning initial coordinates of the target robot, traversing end point coordinates from the production task, and generating a moving route, wherein the moving route consists of a plurality of grid lines with the same length, at least obtaining a first grid line and a second grid line, and marking out line coordinates of each grid line; And S300, acquiring configuration data of the target robot, calculating expected arrival time of a second grid line based on the position and the length of the grid line when the target robot arrives at the first grid line, generating a lookup table, wherein the lookup table consists of line coordinate items and expected arrival time items of the second grid line, establishing a corresponding relation between the target robot and the lookup table, judging whether the same items exist in all the lookup tables, if so, defining the second grid line corresponding to the same items as a conflict line, searching out an alternative route of the conflict line by utilizing a nearest neighbor algorithm, and updating the moving route. Further, the step of defining a working area of the industrial robot, selecting a deployment position of a plurality of identification lamps in each industrial robot, and setting a color of each identification lamp includes: Formulating a numbering rule, and determining the number of each industrial robot; and integrating the numbers and the combined codes to generate a comparison table. Further, the step of generating the combined code includes: the flicker frequency of each identification lamp is configured, the flicker time is recorded, and the position is dynamically updated; the production tasks are clustered into several types and the combined codes are adjusted. Further, the step of meshing the working area, selecting a reference point, constructing a coordinate system, and receiving a production task uploaded by a preset terminal includes: setting the priority of each production task, wherein the priority at least comprises high, medium and low; identifying the priority corresponding to each identical item, sorting the identical items according to the order of the priority from high to low, and defining the first identical item of the sorting as a priority item; And opening the use authority of the conflict line to the priority item. Further, the step of locating the initial coordinates of the target robot, traversing the final coordinates from the production task, and generating a moving route includes: defining two endpoints of the grid line as grid points; based