JPH0488509A - Locus control method for robot - Google Patents

Abstract

PURPOSE:To perform every kind of continuous action of a robot by enabling interpolation to be applied on both a position and posture. CONSTITUTION:It is assumed that preceding action is an operation A and succeeding action is an operation B. As for the operation B, re-interpolation from (b) to (d) is applied similarly as to the operation A, and assuming that synthesis starting time is the same(for, example, Tpd), an excessive acceleration/ deceleration value is shown as in (f). Also, assuming that the synthesis start time for the position and the posture are Tpd, Tsd, the synthesis completion time of the position and the posture is shifted by ¦Tpd-Tsd¦. Therefore, an acceleration/deceleration pattern for the operation B is generated by shifting the synthesis starting time by ¦Tpd-Tsd¦ that is difference between the deceleration start time Tpd and Tsd of the position and the posture, and the synthesis starting time of the position and the posture are set as Tpd and Tsd, respectively.

Description

DETAILED DESCRIPTION OF THE INVENTION (A) Field of Industrial Application This invention relates to a method for controlling the trajectory of a robot, and particularly to controlling the trajectory of the tip of a robot arm having multiple degrees of freedom.

(b) Problems to be solved by conventional techniques and inventions When controlling the trajectory of the tip of a multi-degree-of-freedom robot arm, position interpolation and posture interpolation are performed from the current position and orientation to the target position and orientation. For posture interpolation, there is a known interpolation method in which a rotation axis is set as the target posture by rotation around the l-axis, and the rotation is rotated around that axis (for example, the 24th
(See Proceedings of the 5th ICF Academic Conference (July 1985), lecture number 2908).

At that time, it is well known how to perform position interpolation and posture interpolation according to the set motion parameters, calculate the required time for each, take the smaller one as the motion time, and repeat the position and posture interpolation again to avoid motion. (For example, see Japanese Patent Laid-Open No. 64-25200).

To explain using figures, Fig. 3(a) and Fig. 3(b)
), respectively. If the acceleration/deceleration pattern of the position and posture is determined from the movement amount and the motion parameters as shown in FIG.
(attitude interpolation end time), the acceleration/deceleration parameters for attitude interpolation are changed as shown in FIG. 3(d). Here, Tpd in the figure is the time from the start of position interpolation to the start of deceleration in position interpolation, and Tsd is the time in posture interpolation.

Here, in order to eliminate unnecessary stopping motions during robot motion, in order to realize continuous motion by connecting two motions continuously and smoothly without stopping, conventionally the next motion was started at the start of deceleration of the previous motion. By starting the motion, a continuous motion trajectory was generated.

However, as seen in, for example, Japanese Unexamined Patent Publication No. 64-26911, this method is only used to smooth the connection in position interpolation. It is used only as a method to say "start acceleration", and only takes into consideration position interpolation, and posture interpolation is not discussed, and the problem is that the connection part in posture interpolation is not smooth. there were.

To give a specific example, when starting the articulation of the two movements at Tpd, which is the start of deceleration in position interpolation, as shown in Fig. 4 (a), the connection part of the position interpolation will start as shown in Fig. 4 (a). ), and there seems to be no problem.However, regarding attitude interpolation, the dotted line in FIG. 4(b) results in excessive acceleration/deceleration values. Note that (Tpd+Tp) in the figure is the total time required for the two operations.

The present invention has been made in consideration of such circumstances, and provides a robot trajectory control method that allows smooth position interpolation and posture interpolation.

(c) Means for Solving the Problems The present invention provides two robot trajectory control methods that interpolate the trajectory from the current position and orientation to the target position and orientation using appropriate acceleration/deceleration patterns. When generating a continuous motion trajectory of a motion, a continuous position motion trajectory is generated from a first position acceleration/deceleration pattern and a second position acceleration/deceleration pattern, and corresponds to the first and second position acceleration/deceleration patterns, respectively. A continuous posture movement trajectory is generated from the first posture acceleration/deceleration pattern and the second posture acceleration/deceleration pattern, and the first position and posture acceleration/deceleration pattern start accelerating at the same time, and the first position acceleration/deceleration pattern starts accelerating. The second position acceleration/deceleration pattern starts accelerating when deceleration starts, the second position acceleration/deceleration pattern starts accelerating when the first attitude acceleration/deceleration pattern starts decelerating, and the first position and attitude acceleration/deceleration pattern start accelerating. This robot trajectory control method is characterized in that the patterns end at the same time, and the second position and posture acceleration/deceleration patterns end at the same time.

(D) Effect When combining two movements, the next movement is started when the previous movement decelerates, not only for position interpolation but also for posture interpolation.
By generating acceleration/deceleration patterns so that each interpolation is completed at the same time, it is possible to achieve smooth connection in both position interpolation and posture interpolation.

(e) Examples Hereinafter, the present invention will be described in detail based on examples shown in the drawings. This invention is not limited by this.

Among two consecutive operations, the previous operation is assumed to be the ``movement'', and the next operation is assumed to be the ``B'' operation. Figure 3(b) shows the posture of A movement.
Then, re-interpolation is performed as shown in FIG. 3(d), and the operation is started. By the way, for operation B, if re-interpolation is performed from FIG. 3(b) to FIG.
), the acceleration/deceleration values become excessive. Also, if the synthesis start time is Tpd and Tsd for position and orientation, respectively,
The position and orientation synthesis end time is shifted by 1Tpd-Tscll. Therefore, for the B motion, an acceleration/deceleration pattern is generated by shifting the position and posture deceleration start times Tpd and Tsd in the A motion by 1Tpd-Tsdi, and each synthesis start time for the position and posture is set to 'rpc+ and Tsd.

Specifically, as for the B operation, Fig. 1(b)
1(d), and calculate the deceleration start times Tpd and Ts of each acceleration/deceleration pattern of the position and posture of the A motion.
d deviation (in this case, since Tsd>Tpd, Tsd−T
pd). Furthermore, the composite operation start time differs depending on the position and orientation, and is set to Tpd and Tsd, respectively.

As a result, the synthesized position and attitude acceleration/deceleration patterns become as shown in Figures 2(a) and 2(b), respectively, and the connection part is excessively large in the attitude interpolation that occurred in Figure 4(b). There will be no acceleration/deceleration values, a smooth connection, and a smooth continuous motion trajectory.

(f) Effects of the Invention According to this invention, it is possible to generate a smooth continuous trajectory of two movements between both the position and posture lines, so any continuous movement of the robot can be made smoother. .

[Brief explanation of the drawing]

FIGS. 1(a) to 1(d) are explanatory diagrams showing acceleration/deceleration patterns of the position and posture of an embodiment of the present invention, and FIG.
a) and FIG. 2(b) are explanatory diagrams showing the acceleration/deceleration patterns of the synthesized position and posture in one embodiment of the present invention, and FIG. 3(a) to FIG. Figure 1 (a
) to Fig. 1(d) corresponding diagram, Fig. 4(a) and Fig. 4(
b) is a diagram corresponding to FIGS. 2(a) and 2(b) of the conventional example. Tp...End time of position acceleration/deceleration pattern, Tp
d...Deceleration start time of position acceleration/deceleration pattern, T
sd...Deceleration start time of posture acceleration/deceleration pattern.

Claims (1)

[Claims] 1. In a robot trajectory control method that interpolates the trajectory from the current position and orientation to the target position and orientation using appropriate acceleration/deceleration patterns, when generating a continuous motion trajectory for two motions, A continuous position motion trajectory is generated from a first position acceleration/deceleration pattern and a second position acceleration/deceleration pattern, and a first posture acceleration/deceleration pattern and a second posture acceleration/deceleration pattern corresponding to the first and second position acceleration/deceleration patterns are generated. A continuous posture movement trajectory is generated from the posture deceleration pattern, and the first position and posture acceleration/deceleration patterns start accelerating at the same time, and the second position acceleration/deceleration pattern accelerates when the first position acceleration/deceleration pattern starts decelerating. , the second attitude acceleration/deceleration pattern starts accelerating when the first attitude acceleration/deceleration pattern starts decelerating, and the first position and attitude acceleration/deceleration pattern ends simultaneously, and the second attitude acceleration/deceleration pattern starts accelerating. A robot trajectory control method characterized in that acceleration and deceleration patterns are terminated at the same time.