Orthometric Height Improvement in Tainan City using RTK GPS and Local Geoid Corrector Surface Models

This paper presents an adaptive learning method for data fusion in autonomous\ndriving vehicles. The localization is based on the integration of Inertial\nMeasurement Unit (IMU) with two Real-Time Kinematic (RTK) Global Positioning\nSystem (GPS) units in an adaptive Kalman filter (KF). The observability\nanalysis reveals that $i$) integration of a single GPS with IMU does not\nconstitute an observable system; $ii$) integration of two GPS units with IMU\nresults in a locally observable system provided that the line connecting two\nGPS antennas is not parallel with the vector of the measured acceleration,\ni.e., the sum of inertial and gravitational accelerations. The latter case\nmakes it possible to compensate for the error in the estimated orientation due\nto gyro drift and its bias without needing additional instruments for absolute\norientation measurements, e.g., a magnetic compass. Moreover, in order to cope\nwith the fact that GPS systems sometimes lose their signal and receive\ninaccurate position data, the self-tuning filter estimates the covariance\nmatrix associated with the GPS measurement noise. This allows the KF to\nincorporate GPS measurements in the data fusion process heavily only when the\ninformation received by GPS becomes reliably available.\n

Real-time kinematic (RTK) global positioning system (GPS) receivers are being marketed for applications where accurate estimates of elevation are required. Because this technology is fairly expensive, the quality of these data should be considered with respect to that of USGS digital elevation DEMs (digital elevation models) that are freely available for most of the United States at the 30-m grid scale and for some areas on 10-m grids. The objective of this study was to compare terrain attributes calculated from USGS DEMs with those calculated from DEMs developed with precision RTK GPS measurements. At two locations in central Kentucky, intensive elevation measurements were collected with RTK GPS. The USGS and RTK data for these areas were interpolated to create 4-m DEMs for comparing terrain attributes. Correlations between the attributes developed from different DEMs were substantially larger for slope and aspect than for curvature, topographic wetness index, and stream power index values. Although there were large quantitative differences in terrain attribute values, the general spatial patterns were similar for the attributes calculated from the RTK and USGS DEMs. However, landscape features defined using DEMs developed with RTK data were more similar to those observed with 10- than 30-m USGS DEMs. For some course-scale applications, 30-m USGS DEMs may be adequate for identifying important landscape characteristics. However, the resolution of many of these features will be substantially better with 10-m USGS DEMs. However, quantitative estimates of terrain attributes at a specific location may be grossly incorrect. For situations where accuracy is essential, precision elevation surveys will be required.

RTK GPS mapping of transplanted row crops

In this paper, we present a complete set of solutions to detect the location of the autonomous horticulture rover used for precision agriculture purposes during the early growth stage of the crops. The horticulture rover autonomously traverses the furrow by aligning itself with the centerline using the dynamic inversion navigation algorithm. The navigation algorithm takes the cross-track error and heading error of the rover deviating from the centerline as the location input, and outputs the desired left and right track speeds to the control system to correct the rover course. The rover has been using the real time kinematic (RTK) -based global position system (GPS) as the sensor to detect the location of the rover, but satellite signals can suffer from sporadic loss, and the geographical characteristics of the environment need to be accurately mapped beforehand. These limits of RTK GPS drastically increase the work load of the operators and decrease the location detection accuracy of the rover. The proposed easy-to-implement machine-vision based location detection solution set accepts the raw pictures taken by the camera mounted on the rover as input and calculate the cross-track and heading errors of the rover as outputs. It neither requires the usage of GPS, nor acquisition of environmental geographical information beforehand. Our test results indicate that the proposed machine-vision based solution set is able to outperform the RTK GPS by providing the cross-track location and heading information of the rover with an error less than 20% of that achieved by a RTK GPS, when the rover is within the furrow during the early growth stage.

Autonomous driving is essential for the automation of agricultural systems, and current research primarily relies on using global positioning system (GPS) signals. However, in environments where GPS signals are unstable or denied, such as canopies and smart farms, GPS-based autonomous driving systems face limitations. Therefore, this study proposes an unmanned aerial vehicle autonomous driving system using three-dimensional light detection and ranging simultaneous localization and mapping (3D LiDAR SLAM). The proposed algorithm follows three steps: (1) A map is generated using 3D LiDAR. Because it targets an environment with height restrictions, lightweight and ground-optimized LiDAR odometry and mapping (LeGO-LOAM), which has the characteristic of extracting the ground, is used as a 3D LiDAR SLAM algorithm. (2) Noise is removed through post-processing. (3) A path is generated using the A* algorithm. To evaluate the performance of the proposed system, field evaluations were conducted in an outdoor canopy orchard using GPS and real time kinematic (RTK) GPS data as references. The experimental results verified the limitations of existing GPS- and RTK GPS-based methods and demonstrated the good performance and effectiveness of autonomous driving using 3D LiDAR SLAM. These findings suggest the feasibility of automated operations in environments where GPS is unstable or denied, and efficient agricultural automation may be achieved in conjunction with future harvesting and pollination operations.

Network-based Real Time Kinematic (NRTK) GPS positioning is considered to be a superior solution compared to the conventional single reference station based Real Time Kinematic (RTK) GPS positioning technique whose accuracy is highly affected by the distance dependent errors such as satellite orbital and atmospheric biases. NRTK GPS positioning uses raw measurements gathered from a network of Continuously Operating Reference Stations (CORS) in order to generate more reliable error models that can mitigate the distance dependent errors within the area covered by the CORS. This technique has been developed and tested considerably during recent years and the overall performance in terms of achievable accuracies, reliability and mobility is as good as or even better than can be achieved using the conventional RTK GPS positioning technique. Currently, there are several commercial NRTK services around the world. In the United Kingdom (UK), for instance, Leica Geosystems in partnership with Ordnance Survey has been offering a NRTK GPS service since 2006. This service is called SmartNet and it can provide continuous centimetric level of accuracy to its subscribers. However, NRTK GPS positioning is particularly constrained by wireless data link coverage, correction transmission delay and completeness, GPS signal availability, etc., which could downgrade the positioning quality of the NRTK results. The paper presents some preliminary testing results of an investigation of the SmartNet service from the end users' point of view. A snapshot of the service's performance was carried out as part of a recent PhD studentship jointly awarded by the UK's Engineering and Physical Sciences Research Council (EPSRC) and Leica Geosystems (UK) to conduct comprehensive research into NRTK GPS quality control measures at the Institute of Engineering Surveying and Space Geodesy (IESSG), the University of Nottingham. In order to evaluate the service's quality several static and kinematic tests were performed using the same type of equipment and in the same way that the SmartNet subscribers would have used it. Centimetric accuracy was generally attained during both static and kinematic tests. This high accuracy was only affected by some level of unavailability mainly caused by GPS signal blockage. Additionally, the influence of the number of satellites in view, dilution of precision (DOP) and age of corrections (AoC) over the accuracy and stability of the NRTK GPS solution was also investigated during this research and presented in the paper.

Stockpiles are accumulations of materials such as coal or ore stored at specific locations. Accurate stockpile volume measurement is crucial in the mining and logistics industries for inventory management and cost efficiency. Conventional methods like Real-Time Kinematic (RTK) GPS rely on surface coordinate measurements but require numerous points, especially for irregular-shaped objects. Newer technologies like Terrestrial Laser Scanners (TLS) offer high-precision alternatives by capturing thousands of points per second, expediting and enhancing the resolution of volume measurements. This study compares TLS and RTK GPS methods in measuring the volume of andesite rock stockpiles at PT Bara Tabang. This research uses a quantitative approach, involving six Ground Control Points (GCPs) measured using the Sokkia GRX2 RTK GPS on October 24, 2024. TLS generated over 8.6 million point cloud data filtered down to 35,197 points, while RTK GPS yielded 2,276 coordinates. Accuracy testing showed very small RMSE values (RMSEr: 0.008 m; RMSEz: 0.007 m), and both LE90 and CE90 demonstrated 90% confidence within a 0.012 m range. Volume calculation using Surpac software with the cut and fill method showed TLS produced a volume of 18,766 bcm (51,982 tons/m³), while RTK GPS resulted in 18,694 bcm (51,782 tons/m³), with a difference of 72 bcm or 0.211%.These results indicate that both methods offer acceptable accuracy; however, TLS provides greater data density, efficiency, and precision, particularly for complex or large-scale stockpile objects. Therefore, TLS is recommended for high-accuracy volume measurement in mining operations that require efficiency and detailed analysis.

An adaptive turning algorithm for a four-wheel robot tractor in the headland is presented in this paper. The navigation sensors consisted of an inertial measurement unit and a real-time kinematic global positioning system (GPS). An objective function based on weights was used to create the navigation path, connecting by continuous primitives. The asymmetric steering mechanism was then taken into consideration with a vehicle model. To follow the path accurately, the slide movement of the robot and the steering rate were taken into account by estimating the turning radius in real time. In addition, the vehicle model was tuned based on the results of each turn. Therefore, the turning control algorithm was optimized on the basis of the specific conditions in the field. Field experiments showed that the robot tractor approached the next path with an average lateral deviation of 3.9 cm at a speed of 1.2 m/s during a turn. Compared to a conventional turning scheme, the time consumption and turning trajectory were decreased by 17% and 21%, respectively. Keywords: autonomous tractor, path planning, dynamic circle-back turning, switch-back turning, robot tractor, reinforcement learning DOI: 10.25165/j.ijabe.20181106.3605 Citation: Wang H, Noguchi N. Adaptive turning control for an agricultural robot tractor. Int J Agric & Biol Eng, 2018; 11(6): 113–119.

Because spatial yield variability is usually related to topographic features, the availability of an accuratetopographic map should provide an opportunity to improve management of agricultural inputs. The objective of this studywas to develop a topographic map using a submeter accuracy differentialcorrected global positioning system (DGPS) andelevation data collected during routine crop management passes throughout multiple years. Two fields (24 and 31 ha) witha relief of 19 to 23 m were selected for this study. Elevation was obtained on 1s intervals using a RealTimeKinematic (RTK)GPS mounted on a pickup truck and driving passes on 3 to 50m intervals (greater distances for relatively uniform areas).Elevation was also obtained from multiple (six to seven) passes (4 to 14m intervals) with a DGPS mounted on a tractor,sprayer, or combine (1 or 2s intervals). Elevation data were interpolated to a 5m grid. Mean elevation from everycombination of passes with the DGPS was compared to elevation from the RTK GPS using the root mean square error (RMSE).A simple linear regression was used to compare the average elevation from the DGPS (based on all passes) to the RTK map.Threedimensional wireframe surfaces developed from the RTK GPS and DGPS data and several transects across thesesurfaces were used to evaluate the success of using the DGPS data. The RMSE decreased from 0.63 and 0.75 with one pass,for Fields One and Two, respectively, to 0.39 with six and seven passes, respectively. Average elevation, as determined fromsix or seven passes with the DGPS, was linearly related to elevation from the RTK GPS (r2 > 0.99). Although absoluteelevation differences were sometimes > 1 m, the DGPS maps were effective in characterizing the dominant topographicfeatures of both fields. Features as small as terraces, which are probably smaller than a typical manageable unit, were evidentin the DGPS topographic map. Agronomically important topographic features were identifiable with a DGPS.

Many applications demand high-precision navigation in urban environments. Two frequency real-time kinematic (RTK) Global Positioning System (GPS) receivers are too expensive for low-cost or consumer-grade projects. As single-frequency GPS receivers are getting less expensive and more capable, more people are utilizing single-frequency RTK GPS techniques to achieve high accuracy in such applications. However, compared with dual-frequency receivers, it is much more difficult to resolve the integer ambiguity vector using single-frequency phase measurements and therefore more difficult to achieve reliable high-precision navigation. This paper presents a real-time sliding-window estimator that tightly integrates differential GPS and an inertial measurement unit to achieve reliable high-precision navigation performance in GPS-challenged urban environments using low-cost single-frequency GPS receivers. Moreover, this paper proposes a novel method to utilize the phase measurements, without resolving the integer ambiguity vector. Experimental results demonstrate real-time position estimation performance at the decimeter level. Furthermore, the novel use of phase measurements improves the robustness of the estimator to pseudorange multipath error.

One of the most important applications of a geoid model is a recovery of orthometric heights from ellipsoidal heights (normally obtained from GNSS). The application of the geoid model for recovering orthometric heights from ellipsoidal heights is normally achieved by fitting the geoid model to a local vertical datum. The fitting procedure is usually accomplished by least squares collocation (LSC), using planar or spherical covariance functions. This procedure warps the gravimetric geoid model onto the local vertical datum, hence the local geoid model derived by this procedure, though convenient for local applications, it is not an equipotential surface. We propose offsets method for practical orthometric height recovery from a geoid model. The proposed procedure is more realistic because it does not constrain the local geoid to be coincident to the local vertical datum. We compare the performance of plannar fitting and offsets methods over Japan using a cross-validation procedure. Results show that offsets method performs better than the normally used planar fitting in the recovery of orthometric heights from ellipsoidal heights using a geoid model. The standard deviations of the differences between established and converted orthometric heights at randomly selected GPS/levelling test points over Japan are ±4 and ±3 cm for planar fitting and offsets methods, respectively. The offsets method is therefore more appropriate for converting ellipsoidal heights to orthometric heights than the planar fitting in the area of study.

Terrain attributes are commonly used to explain the spatial variability of agronomic, pedologic, and hydrologic variables. The terrain attributes studied here (elevation, slope, aspect, and curvature) are estimated readily from digital elevation models (DEMs), but questions remain about how the accuracy and sample spacing of the elevation data affect the estimated attributes. The main objective of this study was to quantify differences in each terrain attribute due to factors affecting DEM accuracy and grid cell size. Three data sources were compared: (i) real‐time kinematic global positioning system (RTKGPS); (ii) satellite‐differentially corrected global positioning system (DGPS); and (iii) U.S. Geological Survey (USGS) 30‐m DEMs. The GPS data from three undulating agricultural fields in northeastern Colorado were interpolated onto 5‐, 10‐, 20‐, and 30‐m grid DEMs. The DGPS and USGS DEMs produced similar elevation differences relative to RTKGPS DEMs, but elevation differences in USGS DEMs were more spatially correlated. Estimates of curvature were highly sensitive to DEM differences and the sensitivity of slope, aspect, and curvature estimates decreased as grid cell size increased. The impacts of DEM accuracy and grid cell size were investigated using correlations between wheat ( Triticum aestivum L.) grain yields and estimated terrain attributes. The highest correlation coefficients were obtained using RTKGPS data, and decreasing the sample spacing or grid cell size below 30 m did not consistently improve the correlations. These analyses on agricultural lands indicate the importance of accurate elevation data for detailed terrain analyses on grid cell sizes of 30 m or less.

The potential use of a real-time kinematics (RTK) GPS receiver for seed mapping with a high level of accuracywas investigated. High-accuracy seed mapping can potentially be used in weed control and plant-specific crop management.A four-row Salvo 650 vacuum planter was retrofitted with an RTK GPS receiver. Four seed detector sensors (one per plantingunit) were mounted directly above the planter shoes. These sensors detected the seeds as they fell through seed tubes. Twolow-cost single-board computers were used to acquire data in real time and display it in the tractor cab. The first computerwas interfaced to the RTK GPS unit to determine seed location and forward speed. The second computer was interfaced toa display unit mounted in the cab. The first computer obtained the GPS time and UTM coordinates every second and storedthem with a reference time (time tag). This computer also monitored the seed detector sensors, time-tagged the seeds fromeach unit, and stored the information in memory. The second computer monitored the first computer and reported the plantersperformance through the display unit mounted in the tractor cab. Field tests were conducted to check the performance of thisplanter over two growing seasons. The first year test results indicated a need to control the sensitivity of the seed detectorsensors used for seed detection. The second year tests showed that the improved system performed very well. The differencesbetween the actual plant and the seed map generated by the RTK GPS based planter were in the range of 30 to 38 mm.

Interest in the study of global gravitational models has increased recently all over the world because it is necessary for height datum transformations. Today the International Center for Global Earth Models (ICGEM) provides the largest collection in the world produced through gravitational data from the gravitational satellite’s missions CHAMP, GRACE, and GOCE … Etc. To allow easy access through the internet with its more intelligent technologies, it is one of the International Gravity Organization’s services. While the Global Positioning System (GPS) has become one of the most preferred technologies in engineering surveying, a major dilemma in GPS survey lies in oval-based elevations. At the same time, orthometric heights are commonly used in the engineering field. Therefore, it is necessary to convert the measured heights by satellites assigned to the elliptical surface into orthometric heights and supported to the geoid surface (mean sea level) through an accurate geodetic model. The differences between orthometric measurement heights from DGPS/leveling data (obtained from 57 points in the study area) are increasingly used by professionals geographical information systems(GIS). However, the local determination of Geoid is necessary for better accuracy of the orthometric height from DGPS. This paper aims to introduce a modern technique for determining elevation, avoiding cumbersome and time-consuming spirit leveling operations. Fast vertical positioning can be obtained using DGPS with geoid models. The Root Mean Square Error (RMSE) is ± 0.19 m with high precision of DGPS derived with EGM08; thus, the more it describes the Earth’s gravitational field in more detail.

Height is an important component in three dimensional coordinates and determination of the position of points for any meaningful development. Ellipsoidal heights from GNSS require geoid model which could be global, regional and local for transformation to orthometric height. The absence of a national geoid model in Nigeria remains a great drawback to develop local geoid for local application in place of global geoid models. The study aims to assess the accuracy of polynomial geoid models in orthometric height determination. Differential Global Positioning System (DGPS) observations were carried out to determine ellipsoidal heights of the point while nine and eleven coefficients were used for the geoid and orthometric height modelling. Model A and Model C used 2-D (x, y) positions with nine and eleven parameters while model B used 3-D (x, y, ∆h) positions with nine parameters. The least-squares method was adopted in computing the parameters of the models. Root Mean Square Error (RMSE) was used to assess the accuracy of the models with the RMSE of model A is 14.3 cm, model B is 15.7 cm and model C is 14.5 cm, respectively. The inclusion of height term (∆h) in model B does not improve the accuracy over model A and model C. Model A with the lowest RMSE is hence the better of the three models. One-way ANOVA test conducted at 95% confidence level, however, revealed that the three models did not differ significantly. Model A having lower RMSE is recommended with GPS determined ellipsoidal heights as an alternative to conventional spirit levelling for orthometric height determination within Akure for engineering and environmental applications.