Real-time Kinematic Global Positioning System Research Articles

Sensing System for Real-Time Measurement of Seed Spacing, Depth, and Geo-Location of Corn: A Proof-of-Concept Study

Abstract. Proper seed placement during planting is critical for achieving uniform emergence, which optimizes the crop for maximum yield potential. While uniform plant spacing and seeding depth are often used by corn growers to determine the performance of precision planters, these parameters can be influenced by factors other than machinery, such as seed germination, insects, diseases, and soil properties (e.g., temperature and moisture). Currently, the ideal way to determine planter performance is to manually measure plant spacing and seeding depth. However, this process is both cost- and labor-intensive and prone to human errors. Therefore, this study aimed to develop and test a proof-of-concept sensing and measurement (SAM) system to measure seed spacing and seeding depth and provide the geo-location of each planted seed. The system consisted of a high-speed camera, light section sensor, potentiometer, and survey-grade real-time kinematic (RTK) global positioning system (GPS) unit. Results demonstrated the potential of the proof-of-concept SAM system for measuring seed spacing, seeding depth, and geo-location of corn seeds. Results showed that seed spacing can be calculated using the generated stitched images, achieving a root mean squared error (RMSE) of 0.63 cm and a coefficient of determination (R2) of 0.87 when compared to actual seed spacing measurements. Likewise, 98% of the recorded seeding depths were within the acceptable tolerance of ±10% error. Finally, GPS coordinates were recorded for individual images, which can be used to locate individual seeds and provide detailed information on missing plants (no seeds).HighlightsCalculated seed spacing showed an RMSE of 0.63 cm and an R2 of 0.87.98% of the recorded seeding depth was within the acceptable error of ±10%.GPS coordinates were recorded for individual images. Keywords: High-speed camera, Image mosaic, Light section sensor, Proof of concept, Seeding depth, Seed spacing.

Assessing climate change associated sea-level rise impacts on sea turtle nesting beaches using drones, photogrammetry and a novel GPS system.

Climate change associated sea-level rise (SLR) is expected to have profound impacts on coastal areas, affecting many species, including sea turtles which depend on these habitats for egg incubation. Being able to accurately model beach topography using digital terrain models (DTMs) is therefore crucial to project SLR impacts and develop effective conservation strategies. Traditional survey methods are typically low-cost with low accuracy or high-cost with high accuracy. We present a novel combination of drone-based photogrammetry and a low-cost and portable real-time kinematic (RTK) GPS to create DTMs which are highly accurate (<10cm error) and visually realistic. This methodology is ideal for surveying coastal sites, can be broadly applied to other species and habitats, and is a relevant tool in supporting the development of Specially Protected Areas. Here, we applied this method as a case-study to project three SLR scenarios (0.48, 0.63 and 1.20m) and assess the future vulnerability and viability of a key nesting habitat for sympatric loggerhead (Caretta caretta) and green turtle (Chelonia mydas) at a key rookery in the Mediterranean. We combined the DTM with 5years of nest survey data describing location and clutch depth, to identify (a) regions with highest nest densities, (b) nest elevation by species and beach, and (c) estimated proportion of nests inundated under each SLR scenario. On average, green turtles nested at higher elevations than loggerheads (1.8m vs. 1.32m, respectively). However, because green turtles dig deeper nests than loggerheads (0.76m vs. 0.50m, respectively), these were at similar risk of inundation. For a SLR of 1.2m, we estimated a loss of 67.3% for loggerhead turtle nests and 59.1% for green turtle nests. Existing natural and artificial barriers may affect the ability of these nesting habitats to remain suitable for nesting through beach migration.

Adaptive turning control for an agricultural robot tractor

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.

Accuracy Comparison between GPS Real Time Kinematic (RTK) Method and Total Station to Determine The Coordinate of An Area

Abstract. Survey with GPS Real Time Kinematic (RTK) has the advantage of being faster and easier than the total station, but on the other hand the accuracy of GPS Real Time Kinematic (RTK) is considered lacking. This study was to determine the comparison of accuracy and efficiency of measuring land parcels using a total station and GPS Real Time Kinematic (RTK) method. The research location is at the Universitas Negeri Semarang campus by selecting areas that are open or unobstructed to satellites and congested areas or which have many obstacles to satellites. The results of this study indicate that for open areas, measurement with GPS Real Time Kinematic (RTK) method reaches a horizontal accuracy of 0.040 m with a time of 16 minutes 16 seconds. While the measurement using a horizontal accuracy of 0.00 Total Station with a length of time of 26 minutes 47 seconds. For areas that are densely measured, GPS Real Time Kinematic (RTK) achieves horizontal accuracy of 10.053 m with a length of time of 39 minutes 27 seconds. While the measurement using a precision horizontal Total Station 0.00 with the length of time 25 minutes 41 seconds.

Sprint diagnostic with GPS and inertial sensor fusion

The purpose of this research was to develop a wearable, low-cost prototype based on real-time kinematic GPS and a microelectromechanical inertial measurement unit to measure the sprinting velocity of an athlete. The software package RTKLIB was used to calculate the RTK-GPS positions and different Kalman filters were implemented to provide a loosely coupled sensor fusion. With this setup, we performed empirical studies to determine whether the velocities obtained by this novel approach are sufficiently accurate for a performance orientated training. Therefore, field tests for 30- to 400-m sprint distance were conducted with simultaneous measurements with different reference systems, such as a laser device or timing gates. The evaluation revealed a correspondence between prototype and reference systems with distance and timing errors of $$\pm \, 2\,\%$$ and high correlations for the velocities (R = 0.996, P <0.001) for 68 % of the trials. However, for remaining 32 % of the trials no acceptable performance parameters could be obtained due to GPS problems. Overall, the developed prototype showed great potential and might allow closing the gap between the accuracy and flexibility of the established reference systems as soon as its susceptibility to GPS problems is lowered.

Application of a 3D tractor-driving simulator for slip estimation-based path-tracking control of auto-guided tillage operation

Successful adoption of auto-steering agricultural vehicles in Korean paddy fields depends on the ability to accurately follow a full path, including straight and curved paths for agricultural tasks and headland turning in each row, in the presence of sliding caused by wet ground conditions in relatively small fields. To improve the performance of an autonomous tillage tractor originally developed in our previous study, this article describes a simulation study, conducted using a 3D computer simulator, that accounts for the slippery motion of a virtual tractor on ground with varying adherence properties, In addition, this study performed field evaluation of an auto-guided tillage tractor equipped with a slip estimation-based path-tracking algorithm that is robust to slippage in paddy fields. The test platform was built with a 60-kW tractor equipped with an RTK-GPS (real-time kinematic-global positioning system) and IMU (inertial measurement unit) system, a navigation controller that could estimate the sideslips of the tractor in real-time, and a three-point hitch dynamometer that could measure tillage drafts. The results of the computer simulation confirmed that the slip estimation-based path-tracking algorithm was superior in guiding the tractor along curved paths on the ground with relatively low coefficients of cornering stiffness by reducing the lateral deviations as compared with those obtained with the conventional look-ahead distance method. In field tests run in an arable field, the autonomous tillage tractor equipped with the slip observer algorithm demonstrated improved performance as compared to the previously developed system by reducing the RMSE (root mean square error) for lateral deviation on curved paths from 29 cm to 15 cm.

Exposure of the road network to direct sunlight: a spatiotemporal analysis using GIS and spatial video

ABSTRACTThe direct sunlight hides risks when a vehicle is moving on the roads during summer, as the position of the sun can block a large amount of drivers’ visibility. This paper presents a methodology for validating/calibrating the primary results of a GIS-based spatiotemporal analysis of direct sunlight throughout a rural road with the use of spatial video. The study area is Rethimno, Greece, and the data used for this simulation is the topography, the road network and the sun position for a specific time during summer. The results of the simulation are illustrated on a risk map which indicates segments of roads at high risk in terms of direct sunlight. Moreover, the results have been tested, validated and calibrated in situ. A specialized car with a video camera and a Real Time Kinematic GPS on board was used to evaluate the primary results of the simulation. The ability of the spatial video to validate GIS-based modeling of moving vehicles is a valuable result of this work. Applying the analysis for long time periods, may lead to prevention policies adoption related to accidents of direct exposure to sunlight. Furthermore, this methodology could be an additional module in car navigation systems.

Compaction quality assessment of rockfill materials using roller-integrated acoustic wave detection technique

Abstract During conventional earth-rock dam construction, quality control and acceptance (QC/QA) are based on limited spot tests of material density at random locations which may not be representative of the compacted area and may consist of potential bias. Based on roller-integrated acoustic wave detection technique with real-time kinematic global positioning systems, this research adopted sound compaction value (SCV) as a real-time monitoring index for the dam compaction quality, and subsequently proposed an SCV-based assessment method to estimate the compaction quality of rockfill (RF) materials. Additionally, a geostatistical method (Kriging) was adopted to obtain estimates for both SCVs and compaction quality at any location on the work area, thereby calculating the qualified rate of compaction quality for the entire work area. A case study on the Qianping project in China indicates a strong linear correlation between the SCV and the compaction parameters as well as the dry density of RF materials; therefore, it can serve as a reliable index for monitoring compaction quality of RF materials. Assessment of compaction quality on the entire work area can be quickly and continuously achieved using the proposed method, which can effectively overcome the above limitation of conventional testing, timely feedback compaction information to avoid quality defects, and availably improve the construction quality of earth-rock dams.

Development and application of a strawberry yield-monitoring picking cart

Strawberries in California have a $2 billion direct economic impact on the state; however, they are currently produced based only on uniform field management techniques. Creating yield maps of strawberries could allow for variable-rate and site-specific applications of inputs, which could improve productivity and reduce environmental pollution. This paper presents the development and application of an instrumented strawberry-picking cart for yield mapping. During manual harvest of strawberries planted on raised beds, pickers walk inside the furrows, pick fruit from the beds on both sides, and deposit them into a tray placed on a picking cart. A ‘smart’ picking cart, similar to the standard carts, has been designed and instrumented with several types of sensors including load cells, a real-time kinematic global positioning system (RTK GPS), a microcontroller, and an inertial measurement unit (IMU). This instrumented cart serves two purposes: to work in sync with tray-transporting robots during robot-aided strawberry harvesting, and to create yield maps of strawberry fields. A yield map for an approximately 300m2 plot of a strawberry field in Salinas, California, was generated after the cart was calibrated. During the yield-monitoring experiment, 13.5 trays, each with a capacity of about 4.2kg of strawberries, were filled with fruit. The mean prediction accuracy of the mass of full trays measured by the load cells was calculated to be 4.8%.

Spit and Inlet Morphodynamics of a Tropical Coastal Lagoon

The morphological changes of spits and inlets of the Chilika lagoon, the largest brackish water tropical coastal lagoon in Asia, are investigated using real-time kinematic GPS observation and numerical models during 2009–2013. The seasonal/interannual variations of the spit and inlet cross-sectional areas with varying widths and depths are recorded in association with different physical processes. The results show significant changes in spit morphology: particularly, the south spit accreted continuously, while the middle and north spits eroded. The cross-sectional depth of inlets becomes narrower and deeper during summer and winter seasons, while they are wider and shallower during the monsoon. The model results show that sediment transport rate is larger during monsoon and summer, while it is relatively less during the winter. Alongshore, sediment transport is predominantly northward throughout the study period. The result shows that gain/loss of the spits and closure/opening of inlets are significantly controlled by the high wave power, longshore drifts, and river discharge. The study demonstrates that the combined use of observational and numerical models is very effective to understand the changes of spit and inlet morphology and their impact on ecological conditions of the lagoon environment.

Bank erosion processes measured with UAV-SfM along complex banklines of a straight mid-sized river reach

Abstract. We apply structure from motion (SfM) photogrammetry with imagery from an unmanned aerial vehicle (UAV) to measure bank erosion processes along a mid-sized river reach. This technique offers a unique set of characteristics compared to previously used methods to monitor banks, such as high resolution and relatively fast deployment in the field. We analyse the retreat of a 1.2 km restored bank of the Meuse River which has complex vertical scarps laying on a straight reach, features that present specific challenges to the UAV-SfM application. We surveyed eight times within a year with a simple approach, combining different photograph perspectives and overlaps to identify an effective UAV flight. The accuracy of the digital surface models (DSMs) was evaluated with real-time kinematic (RTK) GPS points and airborne laser scanning of the whole reach. An oblique perspective with eight photo overlaps and 20 m of cross-sectional ground-control point distribution was sufficient to achieve the relative precision to observation distance of ∼1 : 1400 and 3 cm root mean square error (RMSE), complying with the required accuracy. A complementary nadiral view increased coverage behind bank toe vegetation. Sequential DSMs captured signatures of the erosion cycle such as mass failures, slump-block deposition, and bank undermining. Although UAV-SfM requires low water levels and banks without dense vegetation as many other techniques, it is a fast-in-the-field alternative to survey reach-scale riverbanks in sufficient resolution and accuracy to quantify bank retreat and identify morphological features of the bank failure and erosion processes. Improvements to the adopted approach are recommended to achieve higher accuracies.

Estimation of earth-slide displacement from GPS-based surface-structure geometry reconstruction

We propose a simple method to estimate the surface displacement of earth slides. It is based on the assumptions that earth slides move by sliding along a discrete basal-slip surface and extensional structures, which commonly characterize their depletion zone, completely accommodate the occurred displacement. The proposed method is based on the reconstruction of the cross-sectional geometry of structures crossed by an arbitrary chosen longitudinal profile, aligned with the direction of the movement. Structure geometry reconstruction is completed using high-precision GPS survey and the reconstructed geometry is then used as basis to measure horizontal and vertical components of the displacement accommodated by each single extensional structure. Total horizontal and vertical earth-slide displacements are calculated as the sum of displacements accumulated at each single structure. We test the method at the Summonte earth slide in southern Italy using real-time kinematic GPS field mapping. Results from our estimation were validated considering a single displacement vector derived from the direct and indirect measurement of the position of an isolated tree located below the extensional zone of the slide.

Parallel point-to-point tracking for agricultural Wide-Span Implement Carrier (WSIC)

An automatic guidance and control system for the stationary operating mode of an agricultural Wide-Span Implement Carrier (WSIC) was developed and tested on a robotic experimental platform. Control algorithms for the parallel point-to-point tracking problem (forward point tracking and periodic operation process) were designed. Two robots were instrumented with a dual-rover Real-Time Kinematic Global Positioning System, inertial measurement units, XBee-PRO wireless communication modules, and control processors. The experimental platform executes the control sequence in a master-slave cooperative manner. Results of the single point tracking experiments show that the designed velocity and steering angle control laws can effectively guide the robots from a start point to a target. The average and root-mean-square of the offset errors were less than 0.04 m for both the 2-m and 3-m implement width experiments. The lateral and orientation errors were less than 0.16 m and 11° respectively. The sequential point-to-point tracking experiments confirmed the overall effectiveness of the point tracking control and the master-slave cooperative control. The motions of the master and slave robots were in good synchronization, which can permit high efficiency during autonomous WSIC operations.

Suitability of open-access elevation models for micro-scale watershed planning.

Watershed planning is a major issue in Turkey and other parts of the world. Surrounded by seawater on almost three-quarters of its international borders and by sheer mountains along the coastal regions and throughout the country, Turkey experiences a range of climatic changes, which constantly shape its topography. Recently, the occurrences of floods, landslides, and torrents have increased, forcing decision-makers to come up with solutions to manage and rehabilitate the upper watersheds in order to stop or limit the impact of disasters on downstream areas. Possible solutions should reduce flow coefficients, erosion, and sedimentation and increase reservoir capacities. It is expected that torrent volumes will decrease, drainage regimes on slopes will be better organized and adjusted, thawing snow will be better deposited and delayed, evapotranspiration will increase, surface runoffs will be delayed, and water regimes will be better managed, meaning that flood and torrent control will be achieved. For the reasons mentioned above, watershed parameters need to be firmly set. In the scope of this study, the elevation, slope acreage, and reservoir capacity of a small watershed, as extracted from open-access elevation models, were compared to a real-time kinematic (RTK) global positioning system (GPS)-generated point cloud and the resulting elevation model through various geospatial and analytical means. The Shuttle Radar Topography Mission (SRTM) C-band digital elevation model (DEM) (version 3) proved to be a satisfactory method in making residual, correlation, mean, and reservoir capacity comparisons. An L-band Advanced Land Observing Satellite (ALOS) phased-array-type synthetic aperture radar (PALSAR) and an X-band DLR_SRTM ASTER were slightly superior methods in terms of defining a greater number of slope categories than the other models. Finally, DLR_SRTM and SRTMv3 could match a greater number of slope façades than the other models. Seventeen years after its acquisition, SRTM and its derivatives have continued leading the topographic definition of the Earth.

Monitoring river morphology & bank erosion using UAV imagery – A case study of the river Buëch, Hautes-Alpes, France

River morphology dynamics and river bank erosion were mapped using multi-temporal images acquired in June 2014 and June 2015 by an Unmanned Airborne Vehicle (UAV). The selected study sites are located in two dynamic parts of the floodplain of the river Buëch in the Hautes-Alpes province in south-eastern France. The images were processed using the Structure from Motion algorithm into high spatial resolution OrthoMosaics of 5 cm pixel size and DEMs (Digital Elevation Model) of 10 cm pixel size. The positional and vertical accuracy of the UAV products were evaluated using Real Time Kinematic GPS (RTK-GPS) points measurements of markers laid out in the floodplain during image acquisition. Obtained accuracies are centimeters to decimeters. River morphology such as channel displacements, gravel bank displacement and avulsions were evaluated using the OrthoMosaics. The Buëch river shows mainly braided river properties but at locations some meandering river properties were observed. Bank erosion volume calculations were made by comparing the high spatial resolution DEMs of 2014 and 2015. Accuracy of bank erosion assessment was evaluated using RTK-GPS transects of known sites of bank erosion. Bank retreat could be mapped at centimeter to decimeter detail but sometimes hampered by overhanging vegetation, water glitter and shadows. Our conclusions are that time-series of high spatial resolution UAV images can be acquired in a flexible and easy way, the individual images can nowadays be processed in a straightforward way into suitable products i.e. OrthoMosaics and DEMs which are valuable products for land administrators having a responsibility to survey and monitor rivers and control them. Accuracy of the UAV products is high in XYZ direction and sufficient for river monitoring purposes and for designing management measures.

GPS/BDS Medium/Long-Range RTK Constrained with Tropospheric Delay Parameters from NWP Model

Tropospheric delay is a major error source that affects the performance of the Global Navigation Satellite Systems (GNSS) Real Time Kinematic (RTK) positioning especially for the medium/long-range baseline. Although high precision tropospheric delay can be estimated by GNSS carrier phase measurement, together with position and ambiguity, a relatively long period of convergence time is necessary. In this study, we develop a new GPS/BDS RTK algorithm constrained with a tropospheric delay parameters from Numerical weather prediction (NWP) model for medium/long-range baselines. The accuracy of the tropospheric delays derived from NWP is assessed through comparisons with the results of GAMIT (GNSS at MIT). The positioning performance with standard GPS RTK, standard GPS/BDS RTK, the developed NWP-constrained GPS RTK and NWP-constrained GPS/BDS RTK over medium/long-range baselines are compared in terms of the initialization time and the positioning accuracy. Experiment results show that the mean differences between the NWP and GAMIT zenith tropospheric delay (ZTD) are between −5.50 mm and 5.60 mm, and the RMS values of the NWP ZTD residuals are from 24.02 mm to 32.62 mm. A reduction in the initialization time of over 41% and 58% for medium- and long-range baselines can be achieved with the NWP-constrained RTK (both GPS alone and GPS/BDS RTK solutions) compared to the standard RTK solution, respectively. An improvement of over 30% can be found with the GPS/BDS RTK compared with that of the GPS alone RTK for both standard and the NWP-constrained modes. The positioning precision of NWP-constrained GPS/BDS RTK is better than 3 cm in the horizontal direction and better than 5 cm in the vertical direction, which satisfies the requirement of the precise positioning service.

Motion tracking of a wind turbine blade during lifting using RTK-GPS/INS

This paper presents a case of object motion tracking where a real-time kinematic Global Positioning System (RTK-GPS) is used for correction of data obtained from an inertial navigation system (INS). The displacements obtained from the RTK-GPS and the accelerations obtained from the INS are combined using an unscented Kalman filter (UKF). The combined data are used to assess the motion of a 6 MW turbine blade during a hoisting operation and are compared to the blade motion obtained by use of three Leica total stations, that have been used for verification. Special focus in this paper goes to merging of heterogeneous data (displacements, rotations, and accelerations) obtained from different measurement systems. Analysis of the motion data shows that the RTK-GPS/INS measurement system allows for very accurate recording of displacement and rotation measurements.

Evaluation on the performance of single and dual frequency low cost GPS module observation using geodetic antenna

GPS modules have been used for various applications in recent years. Its early development came in parallel with the advancement of Unmanned Aerial Vehicle (UAV) technology. Nowadays, it is also used in in geographic information system (GIS) data acquisition/census, mapping surveys, structure stability monitoring systems and many other applications. GPS modules generally have several positioning features, including standard positioning service (SPS), static positioning, precise point positioning (PPP), post processing kinematic (PPK) and real time kinematic (RTK) GPS. GPS modules in general are only equipped with a microstrip-type antenna or better known as patch antenna. Results from related research show that GPS module with this type of antenna has sub meter accuracy when used for PPK or RTK GPS method. The use of geodetic antennas is very potential to increase GPS position accuracy by up to centimeter level. This paper discusses the evaluation of GPS module measurements with geodetic type antennas for precise positioning using RTK GPS. This paper is focused on the resolution of GPS cycle ambiguity that is often expressed by the term fixing ratio and the accuracy of measurement results obtained. To provide a comprehensive description of the performance of GPS module, in this research two types of GPS module were used; single and dual frequency. Both types of GPS modules were used to conduct simultaneous observation on an open and obstructed observation location.

3D rock slope data acquisition by photogrammetry approach and extraction of geological planes using FACET plugin in CloudCompare

Structure from motion (SfM) permits a fast and inexpensive way to obtain data necessary for geotechnical characterization of rock slope. Unlike scanline survey, a traditional technique, SfM can reduce the risk, time, cost, bias in measurement and improve the accessibility, quantity and quality of data collection. A small and low cost Unmanned Aerial Vehicle (UAV), DJI Phantom 4 Pro mounted with 20 megapixels camera is utilized to capture high quality images in this study with Real-Time Kinematic Global Positioning System (RTK-GPS) being used to obtain coordinates of 10 Ground Control Points (GCP) distributed evenly on the rock slope. Images registered and aligned with GCP render better accuracy results compared to images aligned without GCP. With photogrammetry software, dense point cloud, digital surface model (DSM) and orthophotos can be generated. 3D dense cloud reconstructed in the setting of high accuracy, with millions of points of a rock surface, can be imported into CloudCompare to extract the geological planes using FACET plugin. The major discontinuity sets can be observed and the orientation (dip/dip direction) of the discontinuity sets obtained from the algorithm is accurate and reliable for future geotechnical work such as rock slope stability analysis.

Multi-Temporal Site-Specific Weed Control of Cirsium arvense (L.) Scop. and Rumex crispus L. in Maize and Sugar Beet Using Unmanned Aerial Vehicle Based Mapping

Sensor-based weed mapping in arable fields is a key element for site-specific herbicide management strategies. In this study, we investigated the generation of application maps based on Unmanned Aerial Vehicle imagery and present a site-specific herbicide application using those maps. Field trials for site-specific herbicide applications and multi-temporal image flights were carried out in maize (Zea mays L.) and sugar beet (Beta vulgaris L.) in southern Germany. Real-time kinematic Global Positioning System precision planting information provided the input for determining plant rows in the geocoded aerial images. Vegetation indices combined with generated plant height data were used to detect the patches containing creeping thistle (Cirsium arvense (L.) Scop.) and curled dock (Rumex crispus L.). The computed weed maps showed the presence or absence of the aforementioned weeds on the fields, clustered to 9 m × 9 m grid cells. The precision of the correct classification varied from 96% in maize to 80% in the last sugar beet treatment. The computational underestimation of manual mapped C. arvense and R. cripus patches varied from 1% to 10% respectively. Overall, the developed algorithm performed well, identifying tall perennial weeds for the computation of large-scale herbicide application maps.