JP2021164964A - Robot system - Google Patents

Abstract

To provide a robot system that is able to appropriately correct the position of a robot.SOLUTION: A robot system includes: a robot; a plurality of types of sensors 3, 5 capable of detecting information for calculating coordinates of a tool tip point of the robot; and a processor. The processor generates a point group in the vicinity of a first point and a second point on an operation locus of the tool tip point by an operation program. Each point of the point group has coordinates that the tool tip point moving from the first point toward the second point can reach. The processor sets, at each point of the point group, a weight indicating a probability that may be present, based on the information detected by each sensor 3, 5. The processor groups the point group to extract a group having the highest probability, calculates a position of a gravity center of the extracted group from the coordinates and weights of points belonging to the group, and replaces the coordinates of the second point with the coordinates of the position of the gravity center as new coordinates of the second point when a maximum distance to a point located farthest from the calculated position of the gravity center is equal to or smaller than a predetermined threshold, among the points belonging to the extracted group.SELECTED DRAWING: Figure 2

Description

The present disclosure relates to a robot system.

After teaching the robot offline, a technique is known in which an image is acquired by a camera attached to the robot while the robot is working, and the position and posture of the robot are corrected based on the acquired image (for example). , Patent Document 1).
Since the position of the robot taught offline in an ideal environment fluctuates due to various factors in the actual work environment, the robot can be accurately corrected with respect to the actual work by correcting it using image data. Can be operated.

Japanese Patent No. 4266946

However, for example, when the camera is attached to the hand of the robot, the image is taken from a position close to the work, so that only an image of a partial area of the work can be acquired. In such a situation, when extracting the features of the work surface from the image data, the error in the position of the robot to be estimated becomes large depending on the part of the work to be photographed, and an appropriate correction amount cannot be obtained. There is.

Therefore, there is a demand for a robot system that can appropriately correct the position of the robot even if the error of the position of the robot estimated based on the detection result of the sensor fluctuates.

One aspect of the present disclosure includes a robot, a plurality of types of sensors capable of detecting information for calculating the coordinates of the tool tip point of the robot, and a processor, and the processor is the tool tip point according to an operation program. A point group consisting of a plurality of points is generated in the vicinity of the first point and the second point on the motion locus of the tool, and each of the points in the point group moves from the first point to the second point. Has reachable coordinates, and based on the information detected by each of the sensors, a weight indicating the certainty that can exist is set at each of the points of the point group, and the point group is grouped. , The group with the highest certainty was extracted, the position of the center of gravity of the extracted group was calculated from the coordinates and weights of the points belonging to the group, and the points were calculated from the extracted points belonging to the group. This is a robot system that replaces the coordinates of the center of gravity with new coordinates of the second point when the maximum distance to the point located at the farthest position from the position of the center of gravity is equal to or less than a predetermined threshold value.

It is a perspective view which shows the robot system which concerns on one Embodiment of this disclosure. It is a block diagram which shows the robot system of FIG. It is a flowchart explaining the correction process of the tip point of a tool by the robot system of FIG. It is a flowchart explaining the point group generation routine in the correction process of FIG. It is a flowchart explaining the weighting routine in the correction process of FIG. It is a flowchart explaining the point cloud narrowing routine in the correction process of FIG.

The robot system 1 according to the embodiment of the present disclosure will be described below with reference to the drawings.
As shown in FIG. 1, the robot system 1 according to the present embodiment includes a 6-axis articulated robot 2, a camera (sensor) 3 mounted on the wrist of the robot 2, and a control device for controlling the robot 2. (Processor, memory) 4 and.

The robot 2 has six operating axes driven by a servomotor (not shown), and the rotation angle of each operating axis is detected by an encoder (sensor) 5 provided in the servomotor.
Therefore, the position of the tool tip point TCP of the robot 2 can be calculated based on the rotation angle of each operating axis detected by the encoder 5, and can also be calculated based on the image acquired by the camera 3. can.

The position of the tool tip point TCP based on the image is, for example, the position of the camera 3 with respect to the position of the work by photographing the work with the camera 3, processing the acquired image, and extracting the feature points by a known method. Is calculated, and it is geometrically calculated from the calculated position of the camera 3.

The control device 4 reads the operation program taught by the offline work, operates each axis of the robot 2 according to the operation program, and sets the position of the tool tip point TCP at a predetermined time interval, for example, every 0.1 seconds. Estimate and memorize. As a result, when the next operation program is executed, the operation locus is corrected so that the tool tip point TCP passes through the position estimated at each time point.

More specifically, the control device 4 stores in the memory a first table associated with the coordinates of the tool tip point TCP of the robot 2. In the first table, a second table in which a point cloud is associated with a velocity vector on the operation locus of the tool tip point TCP is stored. The first table and the second table are three-dimensional tables.

The point cloud includes a plurality of points randomly set within the reachable range of the tool tip point TCP while moving from the first point on the operation locus of the tool tip point TCP to the second point after a predetermined time. There is. The coordinates of each point are stored in the second table.

Therefore, as shown in FIG. 3, when the operation program is executed, a point cloud is generated at each time point during execution (step S1). In the generation of the point cloud, as shown in FIG. 4, the coordinates of the current time (second point) arranged on the operation locus of the tool tip point TCP in the operation program and the coordinates of the first point before a predetermined time are used. Is calculated (step S11).

Next, the second table corresponding to the calculated coordinates of the second point is read from the first table (step S12). Further, the velocity vector from the first point to the second point is calculated (step S13), and the point cloud corresponding to the velocity vector is read out from the read second table (step S14). As a result, a point cloud is generated.

Next, a weight is given to each point in the generated point cloud (step S2).
In the weighting step S2, as shown in FIG. 5, first, any of the sensors 3 and 5 is selected (step S21). For example, when the encoder 5 is selected as the sensor, the coordinates of the tool tip point TCP are calculated from the rotation angle of each operating axis detected by the selected encoder 5 (step S22).

Next, one unweighted point is selected (step S23), and the distance L between the coordinates of the selected point and the coordinates of the tool tip point TCP is calculated (step S24). The weight given to each point is inversely proportional to the distance L between the coordinates of the tool tip point TCP calculated from the rotation angle of each operating axis detected by the encoder 5 and the coordinates of each point.

Therefore, the weight W is calculated by dividing the constant A by the distance L (step S25). The calculated weight W is stored in association with the coordinates of each point (step S26).

When the coordinates of the calculated tool tip point TCP include a large error, a constant A smaller than that when the error is small is given. The constant A may be set to the same value for all the sensors 3 and 5, or may be set individually in advance by conducting an experiment for measuring an error for each of the sensors 3 and 5. Further, the constant A may be automatically adjusted by machine learning.

It is determined whether or not the weight W is given to all the points (step S27), and if there are points to which the weight W is not given, the steps from step S23 are repeated.
When the weight W is given to all the points, it is determined whether or not the weight W has been calculated for all the sensors 3 and 5 (step S28).

If there are other sensors 3 and 5 for which the weight W has not been calculated, the other sensors 3 and 5 are selected (step S29), and the steps from step S22 are repeated.
When the camera 3 is selected as another sensor, the coordinates of the tool tip point TCP are calculated from the image acquired by the camera 3 (step S22), and the weight W is given to all the points (step). S23 to S27).

When two sensors, a camera 3 and an encoder 5, are provided, two weights W calculated for each of the sensors 3 and 5 are generated as weights for each point.
The two generated weights W are both stored as weights of each point in association with the coordinates of each point.

Next, the points to which the weight W is given are grouped, and the point group is narrowed down (step S3). Grouping of points is performed as shown in FIG.

First, any one point without a group number is selected (step S31), and a group number is assigned to the selected point (step S32). Next, it is determined whether or not there is another point to which the group number is not assigned within the range of the predetermined distance X from the point to which the group number is assigned (step S33), and all the points existing in the range are determined. Is assigned the same group number (step S34).

Then, the steps from step S33 are repeated. In step S33, when it is determined that there is no other point to which the group number has not been assigned within a predetermined distance X from all the points to which the group number has been assigned, there is a point to which the group number has not been assigned. It is determined whether or not to do so (step S35).

If there is a point to which the group number is not assigned, the group number is changed (step S36), and the steps from step S31 are repeated. When a group number is assigned to all the points, all the weights of the points belonging to each group are summed for each group (step S37).

In this case, if the points have a plurality of weights, all the weights are summed. Then, the other points are erased, leaving the point cloud of the group having the largest total weight (step S38). As a result, the point cloud is narrowed down.

Next, as shown in FIG. 1, the coordinates of the position of the center of gravity of the narrowed down point cloud are calculated based on the coordinates and the weight of each point (step S4). Then, the distance (maximum distance) from the calculated center of gravity position to the farthest point is calculated (step S5), and it is determined whether or not the calculated maximum distance is equal to or less than a predetermined threshold value (step S6).

When the maximum distance is larger than a predetermined threshold value, all the points are erased, and a point cloud is randomly generated inside the sphere centered on the position of the center of gravity and the radius is the maximum distance (step S7). Then, the process from step S2 is repeated.
In step S6, when the calculated maximum distance is equal to or less than a predetermined threshold value, the coordinates are corrected by storing the coordinates of the position of the center of gravity as the coordinates of the second point (step S8).

In this way, by executing the operation program taught offline, the coordinates of the passing point located on the operation locus of the ideal tool tip point TCP are based on the information detected by the plurality of sensors 3 and 5. It is corrected based on the coordinates of the tool tip point TCP calculated by.

According to the robot system 1 according to the present embodiment, the coordinates calculated based on the information detected by the plurality of sensors 3 and 5 also include errors, but the errors are balanced by weighting. There is an advantage that an appropriate correction can be made with careful consideration.

In the present embodiment, the encoder 5 and the camera 3 provided in the servomotor are exemplified as sensors capable of detecting the information for calculating the coordinates of the tool tip point TCP, but the present invention is not limited to these. No. For example, when the robot 2 is equipped with an acceleration sensor, the coordinates of the tool tip point TCP can be calculated by integrating the detected acceleration twice.
When one or more laser sensors are provided to detect the shape of the work, the coordinates of the tool tip point TCP are calculated based on the distance information between the laser sensor and the work detected by each laser sensor. can do.

1 Robot system 2 Robot 3 Camera (sensor)
4 Control device (processor, memory)
5 Encoder (sensor)
X Predetermined distance W Weight TCP tool tip point

Claims (6)

With a robot
A plurality of types of sensors capable of detecting information for calculating the coordinates of the tool tip point of the robot, and
Equipped with a processor
The processor
A point cloud consisting of a plurality of points is generated in the vicinity of the first point and the second point on the operation locus of the tool tip point by the operation program.
Each point in the point cloud has coordinates reachable by the tool tip point from the first point to the second point.
Based on the information detected by each of the sensors, a weight indicating the probability of existence is set at each of the points in the point cloud.
The point cloud is divided into groups, and the group with the highest certainty is extracted.
The position of the center of gravity of the extracted group is calculated from the coordinates and weights of the points belonging to the group.
When the maximum distance to the point located at the farthest position from the calculated center of gravity position among the extracted points belonging to the group is equal to or less than a predetermined threshold, the coordinates of the center of gravity position are set as described above. A robot system that replaces the new coordinates of the second point. The processor
The first aspect of claim 1, wherein when the maximum distance is larger than the threshold value, a point cloud is generated inside a sphere centered on the center of gravity position and the radius is the maximum distance, and the process from setting the weight is repeated. Robot system. A memory for storing a first table that is stored in association with the coordinates of the second point is provided.
The first table stores a plurality of second tables that store the coordinates of the point cloud in association with the velocity vector from the first point to the second point.
The processor
The second table corresponding to the coordinates of the second point is read from the first table stored in the memory, and the coordinates of the point cloud corresponding to the velocity vector are read from the read second table. The robot system according to claim 1, wherein the point cloud is generated thereby. The processor
The method according to any one of claims 1 to 3, wherein a value inversely proportional to the distance from the coordinates to each point is set as the weight of each point for each coordinate of the tool tip point detected by each sensor. The robot system described. The processor
The robot system according to claim 4, wherein the weight is obtained by machine learning for each of the sensors. The processor
Select any one of the points in the point cloud and
Give the selected points a group number and
The same group number is given to the points whose distance from each point to which the group number is given is equal to or less than a predetermined distance, until there are no more points to which the same group number is given.
If there is a point that has not been given a group number, then any one other point that has not been given a group number is selected.
Change the given group number,
The robot system according to any one of claims 1 to 5, wherein the point cloud is grouped by repeating the process of assigning a group number to the selected points.

Images