Search

CN-121973199-A - Robot system for automatic hand-eye error verification, verification method and related device

CN121973199ACN 121973199 ACN121973199 ACN 121973199ACN-121973199-A

Abstract

A robot system, a verification method and a related device for automatic hand-eye error verification are provided, the method comprises the steps of placing a hand-eye verification tool in a working area, collecting point cloud data information of the hand-eye verification tool under a visual field by a vision sensor, transmitting the point cloud data information to an information processing computer, identifying the pose of hand-eye verification tool features by the information processing computer according to the point cloud data information collected by the vision sensor, obtaining feature coordinates under a vision sensor coordinate system and collecting poses of a multi-axis robot coordinate system, randomly generating a plurality of other verification points in the visual field of the vision sensor, planning a plurality of groups of multi-axis robot motion paths required by verification to reach the verification points, acquiring point cloud data collected by the vision sensor at the verification position by the information processing computer, identifying the feature poses of the hand-eye verification tool, converting the feature poses into the multi-axis robot coordinate system according to hand-eye calibration relations, measuring hand-eye errors, and completing verification. The invention can complete verification work simply, accurately and efficiently.

Inventors

  • HUANG RONG
  • Hao huan
  • NIE YONGMING
  • WU YIMING

Assignees

  • 仝人智能科技(淮安)有限公司

Dates

Publication Date
20260505
Application Date
20260123

Claims (10)

  1. 1. A robotic system for automatic verification of hand-eye errors, comprising: A multi-axis robot (102) capable of performing a sport work in a work area; The hand-eye verification tool (104) is provided with asymmetric structural characteristics; The visual sensor (103) is calibrated with the hand-eye relation between the multi-axis robot (102), is arranged on an actuating mechanism of the multi-axis robot (102) and acquires point cloud data information of the hand-eye verification tool (104) under the visual field; The information processing computer (101) is respectively connected with the multi-axis robot (102) and the vision sensor (103), receives the point cloud data information acquired by the vision sensor (103), identifies pose coordinate information of features of the hand-eye verification tool (104) to perform feature calculation, plans a motion path of the multi-axis robot (102) and a measurement position of the vision sensor (103) in any other random poses according to the pose coordinate information of the features in a calibration space of the vision sensor (103), ensures the visible target features of the vision sensor (103) in other acquisition positions, and simultaneously controls the planned path of the multi-axis robot (102).
  2. 2. The robot system for automatic hand-eye error verification according to claim 1, wherein the multi-axis robot (102) has at least three degrees of freedom of movement and the self-execution accuracy is calibrated.
  3. 3. A verification method of a robot system based on automatic hand-eye error verification according to any one of claims 1 to 2, comprising: Placing a hand-eye verification tool (104) in a working area; The vision sensor (103) collects point cloud data information of the hand-eye verification tool (104) under the visual field and transmits the point cloud data information to the information processing computer (101); the information processing computer (101) recognizes the pose of the hand-eye verification tool (104) according to the point cloud data information acquired by the vision sensor (103), and acquires the feature coordinates under the coordinate system of the vision sensor (103) and the acquisition pose of the coordinate system of the multi-axis robot (102); the information processing computer (101) randomly generates a plurality of other verification points in the field of view of the vision sensor (103) according to the pose of the hand-eye verification tool (104); the information processing computer (101) acquires the position and posture coordinates for verification through other postures of the vision sensor (103), and plans a plurality of groups of multi-axis robots (102) motion paths required for verification to reach verification points; The multi-axis robot (102) moves to a verification position of the vision sensor (103) according to the planned path, and the point cloud data information of the hand-eye verification tool (104) under the visual field is collected through the vision sensor (103); the information processing computer (101) acquires point cloud data acquired by the vision sensor (103), identifies the characteristic pose of the hand-eye verification tool (104), converts the characteristic pose coordinates into a multi-axis robot (102) coordinate system according to the hand-eye calibration relation, and measures the deviation of the hand-eye calibration relation between the multi-axis robot (102) and the vision sensor (103) to finish verification.
  4. 4. A verification method according to claim 3, characterized in that after the step of placing the hand-eye verification tool (104) in the work area, the vision sensor (103) on the multi-axis robot (102) is guided to reach the visual position of the verification vision sensor (103) in an arbitrary posture, or the multi-axis robot (102) is moved to the corresponding position according to the recorded history verification position.
  5. 5. A verification method according to claim 3, wherein the information processing computer (101) performs the planning of the motion path of the multi-axis robot (102) under the other postures of the vision sensor (103) by combining the hand-eye calibration relation between the multi-axis robot (102) and the vision sensor (103) and the field information of the vision sensor (103) according to the postures of the features of the hand-eye verification tool (104) and randomly generating other verification points in the field of view of the vision sensor (103).
  6. 6. The method according to claim 5, wherein the number of verification points n is 3 or more.
  7. 7. The method according to claim 5, wherein the step of planning the movement path of the multi-axis robot (102) in other poses of the vision sensor (103) by combining the hand-eye calibration relation between the multi-axis robot (102) and the vision sensor (103) and the field-of-view information of the vision sensor (103) comprises: The information processing computer (101) generates optimal position information of the visual sensor (103) under the current gesture according to the gesture information of the hand-eye verification tool (104) characteristics under the coordinate system of the multi-axis robot (102) and the field information of the visual sensor (103); generating optimal acquisition pose coordinates according to the optimal position information of the vision sensor (103) and pose information of the hand-eye verification tool (104) characteristics; generating n random numbers according to the visual field information of the visual sensor (103), wherein the n random numbers represent the angle information of rotation of the optimal acquisition pose coordinates along different coordinate axes; Rotating the optimal acquisition pose coordinates along coordinate axes in different directions by n random angles to generate n random different pose acquisition vectors; converting n different gesture acquisition vectors into a coordinate system of the multi-axis robot (102), and eliminating unreachable points of the multi-axis robot (102); If the number meets the verification requirement, the verification acquisition pose of the multi-axis robots (102) under n different multi-axis robot (102) coordinate systems is recorded, and the sequencing of the optimal sequence is carried out, so that other poses from the multi-axis robots (102) to the vision sensor (103) are planned to acquire the verification pose coordinates.
  8. 8. A verification method according to claim 3, wherein the information processing computer (101) identifies the pose of the feature of the hand-eye verification tool (104) according to the point cloud data information acquired by the vision sensor (103), and acquires the feature coordinates in the coordinate system of the vision sensor (103) and the acquisition pose of the coordinate system of the multi-axis robot (102), and the method comprises the following steps: removing interference miscellaneous points in the point cloud data by adopting a filtering method; The method for structurally representing and semantically segmenting the point cloud information is adopted, and the structural relation and the representation network only comprising the features to be identified of the hand-eye verification tool (104) are segmented and extracted from the acquired point cloud data information; Generating model parameters of the actually measured hand-eye verification tool (104) according to the structural relation and the characterization network only containing the features to be identified of the hand-eye verification tool (104), and obtaining pose information of the hand-eye verification tool (104) under a coordinate system of a vision sensor (103); Recording pose information of the characteristics of the current hand-eye verification tool (104), and calculating the pose information of the characteristics of the hand-eye verification tool (104) under a coordinate system of the multi-axis robot (102) through a known hand-eye calibration relation and the acquired pose of the multi-axis robot (102).
  9. 9. The method according to claim 3, wherein the step of measuring the deviation of the hand-eye calibration relation between the multi-axis robot (102) and the vision sensor (103) comprises measuring the deviation of the pose of the same name point under the coordinate system of the multi-axis robot (102) under the different pose coordinates of the hand-eye verification tool (104), and measuring the deviation of the hand-eye calibration of the vision sensor (103) and the multi-axis robot (102) under the different poses.
  10. 10. A computer readable storage medium having stored therein at least one instruction for execution by a processor in an electronic device to implement the authentication method of any one of claims 3 to 9.

Description

Robot system for automatic hand-eye error verification, verification method and related device Technical Field The invention belongs to the technical field of intelligent robots with bodies, and particularly relates to a robot system for automatically verifying hand-eye errors, a verification method and a related device. Background In the existing robot system, the relative position relation between a visual sensor and the robot system and the accuracy verification of a hand-eye matrix are usually verified by adopting a manual verification method, a tool point (TCP) installed on the robot is manually moved to an acquisition target point to obtain the coordinate of the point under the robot coordinate system, then the coordinate of the point under the sensor coordinate system is calculated through point cloud acquisition and data processing, the point under the sensor coordinate system is converted into the robot coordinate system through a known hand-eye calibration matrix, and finally the deviation between a manually acquired true point and a theoretical calculation point is compared to verify the hand-eye error. In the process, the working point is required to be observed in place by naked eyes, so that the efficiency is low, the precision is unstable, and the experience of operators is relied on. In the prior art, for example, patent application publication number CN113601514a discloses a robot hand-eye calibration accuracy verification system, which is connected with the tail end of a robot through a fixed pen, and verifies the hand-eye error of the robot according to the image position of a manually movable fixed pen point on a touch screen and through coordinate information fed back by the touch screen. For another example, patent application with publication number CN219582867U discloses a calibration accuracy verification device and system for a robot system, where the accuracy verification device includes a base, a reflecting point component, and an accuracy verification component, the center of the reflecting point coincides with the center of the accuracy verification device, and by identifying the reflecting center point as a standard target position, the error of the hand and the eye is verified by comparing the deviation with the true position. The prior art method adopts special structures such as touch screen points, reflecting points and the like to obtain the characteristic coordinate information of the target object, and the transition path in the process needs to be taught manually every time and moves to a specific coordinate position. In addition, in the verification process, the verification tool meeting the system identification parameters is required to be provided with a specific form or structure instead of any verification tool form, and the process is not intelligent enough. Disclosure of Invention Aiming at the problems in the prior art, the invention provides a robot system, a verification method and a related device for automatic verification of hand-eye errors, which can automatically plan other gesture movement paths and shooting positions required by verification through position identification of a verification tool, and obtain hand-eye calibration deviation between a robot and a vision sensor in different gestures through automatic planning, automatic execution and automatic verification modes, so that verification work can be completed simply, accurately and efficiently. In order to achieve the above purpose, the present invention has the following technical scheme: in a first aspect, a robotic system for automatic hand-eye error verification is provided, comprising: the multi-axis robot can perform movement work in a working area; the hand-eye verification tool has asymmetric structural characteristics; the visual sensor is calibrated with the hand-eye relation between the multi-axis robot and is arranged on an executing mechanism of the multi-axis robot to acquire the point cloud data information of the hand-eye verification tool under the visual field; The information processing computer is respectively connected with the multi-axis robot and the vision sensor, receives the point cloud data information acquired by the vision sensor, identifies pose coordinate information of hand-eye verification tool features, performs feature calculation, plans the motion path of the multi-axis robot and the measurement position of the vision sensor in a vision sensor calibration space according to the pose coordinate information of the features, ensures the visible target features of the vision sensor in other acquisition positions, and simultaneously controls the multi-axis robot to execute the planned path. As a preferable scheme, the freedom of movement of the multi-axis robot is at least three, and the execution precision of the multi-axis robot is calibrated. In a second aspect, a method for verifying a robot system based on the hand-eye error automatic verification is provide