JPH0635524A - Sensor position/attitude and route following control method - Google Patents

Abstract

PURPOSE:To surely catch a route in the viewfield of a sensor by estimating the position and the attitude of the route at a new sensing time point based on the route functions which are sequentially produced based on the discrete information given from the sensor. CONSTITUTION:A route following controller is actuated to collect the route information on a work subject through a sensor and also to sequentially produce the route functions to the reference coordinates of the sensor. Based on these route functions, the position and the attitude of an end effector are controlled and the effector follows a target route 1. At the same time, a route function Pi-1(S) is generated to show a route 6 preceding the (i-1)-th section. Then a route function Pi(S) showing the latest route 5 is generated when the latest sensor data Dn is obtained. Based on this function Pi(S), the end effector advances on the route 1 with change of its position and attitude to perform the next sensing operation with a command of a robot control means.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a sensor position / orientation used for controlling a moving path or attitude of a robot arm or an end effector by a sensor with respect to a work object in an assembly / machining operation of a machine part. The present invention relates to a route tracking control method.

[0002]

2. Description of the Related Art In a conventional robot sensor position / orientation route tracking control system, a sensor is attached to the end of the robot together with the end effector, and the position / orientation information of the route ahead of the end effector is obtained while moving together with the end effector. May be collected. At this time, in order to collect stable information, it is necessary to reliably capture the route within the field of view of the sensor.To that end, the position and orientation of the sensor must be accurately controlled so that they follow the target route. I have to. This is particularly important when the field of view of the sensor is narrow.

One of the conventional techniques for this will be described with reference to the drawings. 2A and 2B show an example of a conventional method for determining the position and orientation of a sensor using past sensor data. FIG. 2A is an explanatory diagram using one-point approximation, and FIG. In the figure, 1 is a target path, 2 is an end effector central axis, 3 is a sensor visual field central axis, 4 is a sensor visual field range, and D1 to Dn are sensor data on the target path 1 detected by the sensor in the past.

Here, it is assumed that the sensor has a mechanism capable of rotating around the end effector central axis 2. As shown in FIG. 2A, the latest sensor data D
It is estimated that the path on the straight line L1 connecting the position indicated by n and the current position of the end effector central axis 2 is continuing,
The sensor is controlled around the end effector central axis 2 using a one-point approximation method so that the sensor visual field central axis 3 comes on the straight line L1.

As shown in FIG. 2B, it is presumed that the route continues on a straight line L2 connecting the positions indicated by the latest sensor data Dn and Dn-1, and the sensor field of view is drawn on the straight line L2 based on this difference. The sensor is controlled around the end effector center axis 2 using the two-point difference approximation method so that the center axis 3 comes. In this example, the target path 1 is included in the sensor visual field range 4, but it includes a considerable error.

[0006]

In the control method as described above, the estimation of the continuing direction of the route is performed by approximating a straight line using the past sensor data value itself or the difference. If the range is narrow, or if the radius of curvature of the route is small and the estimation error is likely to be large, there is a drawback that it becomes difficult to capture the target route within the visual field range of the sensor. In view of the above-mentioned problems of the conventional technique, the present invention controls the position / orientation of the sensor based on a higher-level estimated path, so that the sensor reliably grasps the path in its field of view. It is intended to provide the law.

The above-mentioned problems can be solved by the present invention by adopting the following novel characteristic construction method. That is,
A feature of the present invention is that the path output by a sensor that predictively detects the position and orientation of the end effector target path that is attached to the hand of the robot together with the end effector and is located on the work object in front of the end effector. From the position and orientation data and the position and orientation with respect to the reference coordinates of the sensor, a path function representing the position and orientation of the end effector target path at the reference coordinates is determined and output, and the robot function of the robot is determined based on the path function. In controlling the position and orientation of the end effector, the sensor visual field center axis tracking is performed from the current position on the target path drawn by the path function of the end effector and the distance between the end effector and the sensor. An equation of behavior is calculated, and then the position and orientation on the path that the sensor should take is calculated as described above. Set up a system of equations based on the function,
This is a sensor position / posture path follow-up control method in which the solution is calculated and obtained, and the sensor surely captures the end effector target path in the visual field range.

[0008]

According to the present invention, by taking the above-described method, the position / orientation of the route at the time of newly sensing using the route function that is sequentially generated based on the discrete route information from the sensor. Is estimated and the position and orientation of the sensor are controlled based on the estimated value, and the follow-up behavior of the central axis of the sensor visual field range is expressed by an equation.

The position where the central axis of the sensor field of view should be taken from the solution when the equation and the latest path function are made simultaneous,
The attitude is determined, and the angle around the center axis of the sensor's end effector and the attitude of the sensor are controlled based on this result.Therefore, the central axis of the sensor field of view range is used by using the estimated value of the advanced path obtained from the path function. The target path is accurately controlled and the end effector path is reliably captured.

[0010]

Embodiments of the present invention will be described in detail with reference to the drawings. Figure 1
FIG. 6 is an explanatory diagram of a sensor position / posture path tracking control method according to the present embodiment. In the figure, 5 is the path drawn by the latest path function obtained from the sensor data D1 to Dn, and 6 is the previous section D1.
A path drawn by the path function of Dn-1 and 7 is a three-dimensional solid coordinate system composed of three axes XE, YE, and ZE fixed to the end effector. Note that the same symbols as those in the conventional example of FIG. Here, for simplification of description, it is assumed that the sensor has only the degree of freedom of rotation about the central axis 2 of the end effector, and its posture is not changed.

As specific means (not shown) for realizing this embodiment, the end effector attached to the hand of the robot, the position of the end effector path on the work object (not shown) in front of the end effector, and A sensor for detecting and predicting a posture, a path function determining means for obtaining a path function from the detected data and a position of the sensor with respect to reference coordinates, a robot control means for driving an end effector by inputting the path function, and the robot control means. It is composed of an integrated control system which adopts a path following control device equipped with the robot driven by.

The execution procedure of this embodiment will be described with reference to the drawings. In this embodiment, the route tracking control device is operated so that the route function determining means (not shown) sequentially determines the route function with respect to the reference coordinates of the sensor while the sensor collects the route information on the work object as shown in FIG. The target path 1 is generated, and the position / orientation of the end effector is controlled based on the path function to cause the target path 1 to follow.

At this time, a path function Pi-1 (S) representing the path 6 immediately before the i-1th section is generated, and thereafter, the latest path is obtained when the latest sensor data Dn is obtained. A path function Pi (S) representing the path 5 is generated. Based on this path function Pi (S), the end effector further advances on the target path 1 by changing the position / posture according to a command from the robot control means, and the next Perform sensing.

Since the central axis 3 of the sensor visual field range 4 can take a position on the circumference centered on the central axis 2 of the end effector, its follow-up behavior direction is expressed by a formula, and its drawn line is represented by the sensor visual field range. The center axis 3 following behavior direction line L4 of FIG. For the estimated value of the route 5, from the route information, the route function Pi
Find (S). Control is performed by the sensor by solving the intersection point by simultaneously arranging the movement direction equation of the central axis 3 of the sensor visual field range 4 and the path function Pi (S) equation and selecting the solution on the front side of the end effector from the two solutions. The position of the prediction to be made is obtained.

Therefore, the angle of the sensor about the center axis 2 of the end effector can be calculated from this position data, and the robot control means can give a command to the end effector to perform position / posture path follow-up control. This calculation is easily performed if the sensor outputs route information in the coordinate system 7 in which the center axis 3 following behavior direction line L3 of the sensor visual field range 4 and the route function Pi (S) are fixed to the end effector. be able to.

Further, when the sensor has a degree of freedom for changing the posture, the central axis 3 of the sensor visual field range 4 is a normal value in consideration of the normal direction of the work surface given by the path function. The posture of the sensor can be controlled so as to match the line direction. The degree of overlap between the central axis 3 of the sensor visual field range 4, the sensor visual field range 4 and the target path 1 is accurate, and when compared with the explanatory view of the conventional method shown in FIGS. It is obvious that the example position / posture path tracking control is excellent.

[0017]

As described above, according to the present invention, it is possible to estimate the position / orientation of a route at the time of newly performing sensing, which cannot be handled by the conventional technique, so that a higher-level route estimation value can be used. The position / orientation of the sensor can be controlled, and the sensor can surely capture the route in its field of view, which has excellent utility and usefulness.

[Brief description of drawings]

FIG. 1 is an explanatory diagram of a sensor position / posture path tracking control method according to an embodiment of the present invention.

FIG. 2A is an explanatory diagram of a conventional sensor position / posture path tracking control method using a one-point approximation method, and FIG. 2B is the same explanatory diagram using a two-point difference approximation method.

[Explanation of symbols]

1 ... Target path 2 ... End effector central axis 3 ... Sensor visual field range central axis 4 ... Sensor visual field range D1 to Dn ... Sensor data 5 ... Path drawn by latest path function obtained from sensor data D1 to Dn 6 ... Path drawn by the path function of the previous section D1 to Dn-1 7 ... Coordinate system

Claims (1)

[Claims] 1. A position of a path which is attached to a hand of a robot together with an end effector, and which outputs a sensor for predicting and detecting the position and orientation of the end effector target path on a work object in front of the end effector. And orientation data, and the position and orientation of the sensor with respect to the reference coordinates, a path function representing the position and orientation of the end effector target path at the reference coordinates is determined and output, and the robot of the robot is based on the path function. In controlling the position and orientation of the end effector, the sensor field-of-view range center axis tracking behavior is determined from the current position on the target path drawn by the path function of the end effector and the distance between the end effector and the sensor. And calculate the position and orientation on the path that the sensor should take, And the path function, a simultaneous equation is set up, the solution is calculated and obtained, and the sensor accurately captures the end effector target path in the visual field range. Control method.