1. Introduction
The topic of the paper is important for all geo-activities that are related to surveying land. Globally, in geodesy, we have a control frame and, locally, we have a stated national frame; for current works, we establish networks. Progress in this term was presented in [
1], which displasy the most important information for practitioners and scientists in the “Geodesy and Cartography” field of research. Nowadays, many people use the geo-data, geo-information, or some geo-portals [
2], which are produced with reference to control networks. We have many methods of producing land surveying data where geodetic control points play an essential role [
3]. As we know, geodetic control networks provide an established national spatial reference system [
4]. In modern surveying, the geodetic data are usually realized using space (3D) data obtained from the Global Navigation Satellite Systems (GNSS) methods. Today, in GNSS positioning, six satellite navigation systems are used [
5]: GPS of the United States, GLONASS of the Russian Federation, Galileo of the European Union, Compass/Beidou of China, QZSS of Japan, and IRNSS of India. The systems are used together (two or more) to increase the measurement accuracy [
6,
7,
8]. There are more space-surveying techniques that can realize the three-dimensional geodetic data by determining or calculating the three-dimensional coordinates of surface points. The applications of geodetic control points are placed in topographic mapping [
9], engineering construction measurements [
10], and georeferencing survey of point clouds and raster images [
11,
12]. According to the classical perspective in surveying, we need reference control points to establish or check equipment (tacheometer—instrument or GNSS—receiver). In these cases, when using new sensors in surveying (UAV—photogrammetry or LIDAR—systems), the control points are helpful for guaranteeing an equal accuracy of produced geo-data [
13]. In accordance with national regulation (for example in Poland with Regulation [
14]), the base geodetic control network consists of points of the horizontal and the vertical networks. Nowadays, the coordinates of the horizontal (and 3D network) base geodetic control network points mainly are determined using GNSS surveying and the heights of the points are usually determined using geometric precise leveling. The base geodetic control network points are stations of the continuously operating reference system (CORS in Poland named ASG-EUPOS), which established the three-dimensional geodetic data. In the ASG-EUPOS reference network, the average error of the point’s position should not exceed 0.01 m and the error of the point’s height should not be greater than 0.02 m.
Generally, in geodesy and especially in surveying topics, the classical method of new applications and modern configurations is used. Photogrammetric flights (raids) for land survey are realized using unmanned aerial vehicles (UAV) from low-altitude distances [
15]. Triangulation (directions measurements) may be represented using laser scanning technology [
16]. The idea of GNSS surveying can be equated with the implementation of the geodetic intersection task, because the satellite navigation systems are similar to the land-based systems where the satellites act as the reference points [
17]. Nowadays, the innovation concept of direct land surveying is the smart station survey, including integrated tacheometric measurements using real-time GNSS positioning, e.g., by Leica [
18]. The technology of the smart station surveying was defined by the trends in economic and measurement activity too [
19]. Over the last few decades, some innovations in land surveying have been implemented, for example, in establishing the control networks [
20,
21]. Initially, GPS techniques (currently GNSS technologies) help to eliminate the need for the highest (first) order control point densification and improve the accuracy of control network points [
22,
23]. Nowadays, first order control network points are created in accordance with the EUPOS standard for national CORS; the following assumptions were made: existing EPN and IGS stations have been incorporated into the network of reference stations and the mean distance between stations is 70 km. Moreover, the locations of the reference stations were chosen to ensure convenient conditions for GNSS satellite observations and only precise dual-frequency GNSS receivers have been used.
Typically, the estimation of the accuracy of the network horizontal and vertical points are specified with respect to an appropriate national geodetic datum [
24]. The tasks involving constructing control network points will most often need to know the position’s relationship to the realization of the reference frame [
25]. Constructing control networks is a huge challenge that individual countries and research centers are working on. Previously, the horizontal geodetic control networks were established using triangulation and traversing or using photogrammetry methods, while the vertical geodetic control networks were established using geometric leveling. For example, we know the standards and specifications for geodetic control networks from the United States of America [
26]. Today, Global Navigation Satellite Systems and GNSS methods are generally used to establish both horizontal and vertical control network points. In this process, some tasks must be resolved, including planning of observation [
27], and precise point positioning and calculation in the aspect of the solution of the fixed points [
28,
29,
30].
2. Materials and Methods
During the application of a smart station survey for establishing control network points, three measurement methods were performed. The procedure for point positioning was supported by performing two satellite GNSS measurement methods (fast static and RTK/RTN GNSS positioning) and one integration (hybrid method realized by terrestrial and satellite positioning) surveying method.
The reference materials in the research are coordinates (X, Y, and Z) of nine control points established using GNSS static occupation over three days with three sessions every day [
31]. The static sessions were started at 8 a.m., 10 a.m., and 12 a.m.—in three parts each day, with points measured in different sessions. The control network points were situated in the area of the University of Warmia and Mazury in Olsztyn in various field conditions on the Kortowo campus (
Figure 1).
The points were determined using static occupation over 90 min with the recording interval set to 30 s and a 10-degree elevation mask using three Topcon HyperPro GPS/GLONASS receivers. For post-processing purposes, the reference data were obtained from nearby permanent stations (BART, DZIA, ILAW, KROL, LAMA, and OLST) using the ASG-EUPOS POZGEO D service [
32]. Rigorous adjustment was realized in the procedure dedicated for 3D and GNSS networks with the use of C-GEO software from Softline company [
33], in the coordinate reference system EPSG:2178 (name PL-2000, Poland CS2000 zone 7), and in the geoid-based vertical system PL-KRON86-NH. Reference (model) coordinates for the control network points with accuracy were appointed: horizontal errors m
XY ≤ 0.004 m and vertical m
Z ≤ 0.003 m.
The model data were used for calculating the differences of coordinates after applying reference (r) GNSS static occupation and with the use of an individual (i) measurements scenario (the steps of the proposed approach are presented below in
Figure 2).
After calculating the measurement uncertainties, we may classify the surveying results and proposed measuring equipment for position and height determination in the centimeter range. According to professional practice with real-time GNSS positioning, for typical cases: high-quality engineering surveys in monitoring and construction works, with uncertainties D
XY < 0.01 m and D
Z < 0.02 m; basic engineering surveys and real estate surveying, with uncertainties D
XY < 0.03 m and D
Z < 0.05 m; civil engineering surveys, with uncertainties D
XY and D
Z no greater than 0.20 m; and surveying tasks in natural boundaries, with uncertainties D
XY and D
Z no greater than 0.50 m are all used. Similar national CORS guidelines [
32] in this regard state that the estimated precision of real-time surveying is 0.03 m (horizontally) and 0.05 m (vertically).
The performed measurements were the inspiration for defining the following main aims of this study:
- (1)
Checking the possibilities of replacing full static GNSS surveys using the single receiver measurement.
- (2)
A multi-variant data acquisition for reference point stationing from surveys performed using smart station equipment.
- (3)
Investigation of the used impact’s type of reference corrections on the real-time GNSS positioning of the control network points.
In the study of the method with the use of static surveys with a single receiver from the smart station set, analyses were performed of the coordinates of the control points hat were determined by the static occupation over 30 min with different recording intervals set to 1, 5, 10, 15, and 30 s and a 10-degree elevation mask using Leica GS15 receiver. The detailed analyses were mainly realized for the interval sets recorded with 1 and 5 s; this was due to practical recommendations for short static measurements carried out as part of the national satellite positioning system [
32]. The differences between the coordinates calculated from the reference coordinates from the model static GNSS occupation and the results of the post-processing from the single receiver static survey were analyzed. The analyses considered fifteen scenarios of measurements marked in this key, for example: 1 s G&GL mean 1 s interval of recording GPS and GLONASS signals, 10s G mean 10 s interval of recording only GPS signals, and 30 s GL mean 30 s interval of recording only GLONASS signals.
The multi-variant measurements of control network points were realized as traversing. In the study, analyses of the coordinates of the control points that were determined using total angular–linear measurements and RTK/RTN surveys using the Leica Smart Station set were performed. The angles and distances were surveyed using the Leica TS15 tacheometer; it was a robotic total station that measured with accuracy: directions 1″ (0.0003 gon) and distances 1 mm + 1.5 ppm (in standard mode), while a Leica GS15 receiver was used for the real-time technique surveying (in VRS, MAC and SRS corrections).
In the next study, analyses of the coordinates of the control points that were determined using only RTK/RTN surveys with a Leica GS15 receiver were performed. The real-time technique was realized by using multiple reference stations (VRS or MAC) and single reference station (SRS) corrections. Some points were located in more difficult field conditions, in the so-called “city canyons”, under trees, and near buildings. The Kortowo campus is a wonderful place for the study of activity [
34]. In our geo-engineering activities, we carry out many measurement works on the campus. We designate control network points for direct surveys and checkpoints for remote measurement techniques and their location was also the subject of previous publications by poster or paper [
35,
36].
4. Discussion
The possibility of utilizing the GNSS surveys for object navigation and geodetic measurements are dependent on its positioning accuracy, which is strictly related to the number of satellites forming the constellation [
42]. In accordance with the literature [
43], using two satellite systems (GPS and GLONASS) significantly increases the accuracy of measurements.
The results from determining the coordinates of the control points using the static surveys method with a single receiver stated, for all calculation variants, that the error in the position of the point in the horizontal plane (DXY) did not exceed 0.03 m and, quite abnormally, was twice as large as the error in determining the point in the vertical plane (DZ). In general, the largest differences in coordinates (calculated from reference coordinates) were obtained for the adjustment using only one system (GPS), while the minimum differences in the coordinates were found for the solution using observations from both systems (GPS and GLONASS).
In the considered scenarios of the measurements, the biggest difference for the X coordinate equal to 0.160 m was obtained for point 10030 (30 s GL). However, the smallest difference for the X coordinate equal to 0.000 m was obtained for point 10070 (30 s G&GL). The biggest difference for the Y coordinate equal to 0.170 m was obtained for point 10030 (15 s GL). However, the smallest difference for the Y coordinate equal to 0.000 m was obtained for points: 10000 (5 s G&GL), 18970 (5 s GL, 10 s GL, and 15 s GL), 10020 (15 s G&GL, 15 s G), 18966 (15 s G&GL and 30 s G&GL), and 10070 (15 s G&GL). The biggest difference for the height (for Z coordinate), equal to 0.248 m, was also found for point 10030 (30 s GL). On the other hand, the smallest difference for the Z coordinate equal to 0.000 m was obtained for points: 10020 (30 s G&GL) and 10080 (1 s G&GL).
The results from determining the coordinates of the control points using the multi-variant survey method stated that the highest number of maximum coordinate differences from the reference coordinates were obtained for the total angular–linear and RTK/RTN measurements assumed on the basis of corrections from the virtual reference stations (VRS corrections). On the other hand, the most measurement results close to the reference coordinates were obtained for the multi-variant surveys with corrections from a single station (SRS).
In detail, the biggest difference for the X coordinate equal to −0.018 m was obtained for point 10030 (VRS). However, the smallest difference for the X coordinate equal to 0.002 m was obtained for point 10000 (SRS). The biggest difference for the Y coordinate equal to −0.013 m was obtained for point 10080 (SRS). However, the smallest difference for the Y coordinate equal to 0.000 m was obtained for points: 10020 (SRS), 10030 (MAC and SRS), 10040 (SRS), 18966 (MAC), and 10070 (MAC). The biggest difference for the height (for Z coordinate), equal to −0.038 m, was obtained for point 10080 (VRS). On the other hand, the smallest difference for the Z coordinate equal to −0.001 m was obtained for point 10030 (SRS).
In the results from the method establishing control network points by using RTK/RTN surveys, the greatest number of measurement results close to the reference coordinates were obtained for the MAC corrections and for the height for corrections from a single station (SRS).
The highest number of maximum coordinate differences was obtained for the control network points established on the basis of corrections from the virtual reference stations (VRS corrections), while the number of minimum coordinate differences was very similar for all types of corrections.
In detail, the biggest difference for the X coordinate equal to −0.031 m was obtained for point 10030 (VRS). However, the smallest difference for the X coordinate equal to 0.002 m was obtained for the following points: 18970 (SRS), 18966 (VRS), and 10080 (MAC and VRS). The biggest difference for the Y coordinate equal to 0.024 m was obtained for point 10040 (VRS). However, the smallest difference for the Y coordinate equal to 0.000 m was obtained for points: 10030 (SRS) and 10070 (VRS). The biggest difference for the height (for Z coordinate), equal to −0.059 m, was obtained for point 10010 (VRS). On the other hand, the smallest difference for the Z coordinate equal to −0.005 m was obtained for point 10030 (MAC).
In
Figure 4 and
Figure 5, the frequency of the minimum and maximum differences after applying GNSS occupation using the individual types of corrections are displayed (SRS, MAC, or VRS). The minimal differences in coordinates were practically even for SRS, MAC, and VRS corrections. However, the maximal differences in the coordinates were stated for VRS correction, in contrast to MAC correction, which is in accordance with the literature [
44]. It is also practically correct that the high-quality solution was obtained for positioning with the use of a single reference station (SRS correction).
The performed studies did not show any significant difference in the results of the measurements using the GPS system or the GLONASS system. Of course, with the use of both systems (GPS and GLONASS), the number of satellites in the constellation was bigger and the accuracy of positioning of control network points was, therefore, also higher. It was especially noticeable at the points located in more difficult field conditions (near trees and buildings); these are the control points numbered: 10040, 10070, and 10080. Generally, for the control network points that were determined by 30 min sessions of static occupation, the differences of coordinates (X, Y, and Z) to the reference coordinates were no more than 0.030 m. Some of the results of the presented investigations are not very typical, as they were probably influenced by the time window of the satellite positioning and the approximately north–south orientation of the sequence of measured points. For example, the environments of points 10040 and 10070 are relatively poor, but these points have even better results than the other points (see
Figure 3). In accordance with the literature [
45], the urban environment is characterized by the presence of an excessive number of obstacles that produce a multipath effect in the positioning of the GNSS receiver.
Additional assessments were also used in the paper, for results of GNSS positioning were applied in assessing the expected accuracy [
40]. The uncertainty interval calculated for the horizontal and vertical components together corresponds quite well with those presented in this work of the differences of the coordinates. When they are totally included, these could represent the displacement distance (as a vector obtained from D
X, D
Y, and D
Z in a three-dimensional Cartesian system). In the case of multi-variant surveys (terrestrial and satellite hybrid measurements) the observations were adjusted using a classical approach [
46]. The available publications also present the results of a further integration of satellite positioning and terrestrial measurements [
47,
48].
5. Conclusions
In this paper, the application of a smart station survey for creating control network points was presented. As we know from the United States Patent (No. 5,233,357 and Date: 3 August 1993) Smart Station set offered by Leica, results may be obtained using measurements in a terrestrial surveying system and in satellite navigation system separately or in mixed-use (hybrid) systems [
49].
The calculated differences of the coordinates between the reference coordinates from the model static GNSS occupation, the results of post-processing from single receiver static surveys as well as from RTK/RTN surveys performed using a GNSS receiver, and, finally, multi-variant surveys performed using smart station equipment were considered.
The results for determining coordinates of control points using the multi-variant survey method (tacheometry and real-time GNSS methods) stated that the highest differences of coordinates were no more than 0.02 m in the horizontal plane (uncertainty DXY) and no more than 0.04 m in the vertical plane (uncertainty DZ).
After using real-time GNSS occupation, the differences of the coordinates were no more than 0.03 m in the horizontal plane (D
XY) and no more than 0.06 m in the vertical plane (D
Z). Certainly, the high accuracy of the kinematic measurement using SRS correction is influenced by the fact that the used single reference station was located a short distance from the research area (less than 2 km). The network measurement with MAC correction has an interpolation algorithm, which provides greater possibilities and independence from the need to “trust” to VRS correction [
44].
With the use of multi-variant surveys realized using Smart Station angular–linear and RTK/RTN measurements, a weaker solution was obtained on the basis of the corrections from the virtual reference station (VRS correction). The most measurement results close to the reference coordinates were obtained for the multi-variant surveys with corrections from a single reference station (SRS).
Nevertheless, the research results have shown that the use of Smart Station equipment is very hopeful for measuring urban environments, establishing control network points, and for other purposes (in areas with no full GNSS signals, e.g., near trees and buildings). This is in accordance with the results presented in the research of other authors [
50].