Instrumentation & Measurement Magazine 25-5 - 21

Fig. 5. The data acquisition results. (a) The proposed method. (b)The uniform sampling method.
concentric curves in the uniform sampling method are subsampled
by using (10), and the results are shown in Fig. 4. In
the uniform sampling results, the number of sampling points
is 2401. By using the proposed method for resampling, the
number of sampling points can be reduced to 1422. The two
groups of sampling data are saved as TXT file. In robotstudio
6.07, the sampling point file is read, and the manipulator is
controlled and programmed. The robot drives the laser ranging
sensor to scan and measure in the space plane according to
the planned path. When the sensor moves to the designated
position, the reading zd of the sensor is recorded. The robot
first moves to the center position A0
in the plane and records it,
and then gradually moves to the circular track of the outer ring
for point-by-point measurement. To ensure that each planning
point can be accurately recorded, the measurement interval
of the robot is set to 1s/point. When the robot traverses all the
points to be measured in the reference plane, the measurement
ends. The data of the part's surface is stored according to the
number sequence of the acquisition curve, and the storage format
of each data point is (x,y,zd), where x and y are the x and y
coordinates of the corresponding points in the reference plane
measurement track, and zd is the decimal representation of the
laser ranging sensor reading.
After measuring the whole surface of the workpiece, a
={p1,p2,...,pi,...,pn}, pi ∈ R3
point set P*
(xi,yi,zdi
of points can be obtained.
The real data point distribution of the workpiece's surface
can be obtained by the difference operation between the sampling
point pi
A0(x0,y0,zd0
) in the point set and the center point
). The data acquisition results of the proposed
method are shown in Fig. 5a. The data acquisition results
of uniform sampling method are shown in Fig. 5b. The total
time to complete the profile measurement of workpiece using
the proposed method is 1432 seconds. However, the total
time taken to complete the measurement of the workpiece using
the uniform sampling strategy increased to 2411 seconds.
In the uniform sampling result, the number of sampling
points in each circle is the same, so the number of sampling
points at the bottom of the workpiece is sparse, and the
August 2022
number of sampling points at the top of the workpiece is too
dense. Uniform sampling requires manual planning of the
sampling trajectory radius and sampling point distribution.
Due to the introduction of arc length and feature weight, the
proposed method can adaptively plan the trajectory radius
and sampling point interval. Meanwhile, the distribution of
sampling points in the proposed method is more reasonable
than that of uniform sampling points. In this method, the
number of planning points is less than that of uniform sampling
points, and the time consumed by the measurement
process is greatly reduced. As a result, it is more suitable for
data acquisition and data planning of certain large complex
surface components.
Conclusion
In this study, the difficulties of data acquisition and planning
in reverse engineering of some large-scale components in the
aerospace field have been analyzed. An algorithm of measuring
point planning for large surface workpieces is proposed to
realize intelligent and adaptive planning of measuring points
of the surface to be measured. The complexity analysis model
of the sampling point curve is established, and the measurement
planning method is proposed, based on the theory of
non-uniform rational B-spline fitting and B-spline wavelet decomposition.
What is more, a data acquisition scheme for large
an ellipsoid workpiece is proposed and verified by a simulation
experiment combined with the principle of a section line
scanning measurement. The results show that the proposed
method can adaptively plan the distribution of sampling
points, the distribution of sampling points is more uniform,
and the surface coverage is better.
In conclusion, the measuring point sampling strategy proposed
in this paper can be applied to the data acquisition
process of large-scale surface workpieces with some characteristics.
This method basically meets the requirements of surface
measurement, but it needs further experiments and verification
for the distribution effect of measuring points and data
acquisition results of other shape trajectories.
IEEE Instrumentation & Measurement Magazine
21

Instrumentation & Measurement Magazine 25-5

Table of Contents for the Digital Edition of Instrumentation & Measurement Magazine 25-5