THE USE OF UNMANNED AERIAL SYSTEMS FOR OPERATION AND MAINTENANCE OF IRRIGATION SYSTEMS

Transcription

1 THE USE OF UNMANNED AERIAL SYSTEMS FOR OPERATION AND MAINTENANCE OF IRRIGATION SYSTEMS Jorge Flores-Velázquez 1, W. OjedaBustamante 2, Mauro Iñiguez Covarruvias 3, Ronald OntiverosCapurata 4, and Nahún H. García Villanueva 5 ABSTRACT UAS, Unmanned Aerial Sistems, Remotely-Piloted Aerial Devices, or simply, drones are a tool of increasing use in engineering and a situations where a manned inspection is not possible. With the advent of low-cost equipment with sophisticated computer vision, robotics and geomatic engineering, these devices are able to generate high-resolution images and videos even with low cost cameras. Performance of irrigation infrastructure depends on its maintenance to maintain functionality, operability and to ensure life expectancy, which demand frequent supervision and use of resources. This report describes the potential use of UAS s to help monitor operations and maintenance activities of Mexican irrigation systems. The results indicate that UAS s provide the appropriate platforms for a remote sensorbased at reduced cost and labor, to expedite the supervision and monitoring process of hydraulic infrastructure, such as: opportune detection of water leaks, performance of irrigation service, and advances of maintenance jobs. Keywords: UAV, dron, monitoring and supervision system, image processing, irrigation schemes. 1. INTRODUCTION Irrigated agriculture in Mexico develops in a great variety of weather conditions, soil, technological level and sociocultural factors. Mexico has around 6.5 million hectares of irrigated agriculture which represent 25% of the cultivated surface, out of which, 50% of the value of production is obtained. From to total irrigated surface, 91% is superficial irrigation, and the rest, 560 thousand hectares, represent pressurized irrigation systems (PNH, 2008 Conagua, 2010). However, the limit has now been reached for the agricultural frontier because of the lack of water, for this reason it is has become necessary to increase the productivity of irrigation water, so the maximum amount of soil can be cultivated with the minimum amount of water. In the last few years, technologies have been developed that will allow the efficient and rationed use of water with the purpose of facing the increasing demand of consumption. These technologies have allowed for a better control process of production, as a consequence, in the yield of crops; with an emphasis of the proper management of resources. This use has been made possible in part, due to the 1 Researcher, Instituto Mexicano de Tecnología del Agua (IMTA). Paseo Cuauhnáhuac 8532, Jiutepec, Mor. México. CP ); jorge_flores@tlaloc.imta.mx 2 IrrigationEngineeringDepartment Head, Instituto Mexicano de Tecnología del Agua (IMTA). Paseo Cuauhnáhuac 8532, Jiutepec, Mor. México. CP ); wojeda@tlaloc.imta.mx 3 Researcher, Instituto Mexicano de Tecnología del Agua (IMTA). Paseo Cuauhnáhuac 8532, Jiutepec, Mor. México. CP ); mic@tlaloc.imta.mx 4 Researcher, Instituto Mexicano de Tecnología del Agua (IMTA). Paseo Cuauhnáhuac 8532, Jiutepec, Mor. México. CP ); ron@gmail.com 5 Irrigation&DrainageCoordinator, Instituto Mexicano de Tecnología del Agua (IMTA). Paseo Cuauhnáhuac 8532, Jiutepec, Mor. México. CP ); nahung@tlaloc.imta.mx 1

2 implantation of efficient systems in several stages of the production process, from the storage source, to the net irrigation requirement for the allotment being cultivated, as well as the timely manner of its application. Here we combine hydraulic and agronomical factors, but above all, operation from the user. Make the water reach the plant from the source involves a complex distribution network and derivation devices that must be in function to avoid water waste. Keeping in proper condition this network is understated as complicated, partly because of the dispersion and inaccessibility to the site, but also because of the lack of personnel. The technological development has been fundamental in keeping the net and systems, by including accurate instruments to maintain control. Nevertheless, when no on-line data transmission method exists, monitoring is imperative in transporting the device for visual supervision. Among the possibilities for this supervision, the use of unmanned aerial devices can be found (UAS). Because of the technical, operative and economic advantages that a UAS s implies, in the last decade, several procedures for the analysis of information generated by sensors in UAS s have been developed. Among the implemented techniques, the characterization of pixels stands out (un-mixing the sub-pixel) which is the feature of which the resolution of the image depends (Foody et al., 1997) as well as the geometrical features. Several procedures have been used and defined to approach the analysis of information that a pixel provides. For the implementation of these models, it has been contemplated to use (Atkinson et al., 1997), artificial neural networks (ANN), modeling of the linear mixing method and classification. Other methods to value the area based on the size of the pixel include (Gallego et al., 1993; Gonzalez-Alonso et al., 1997; Thenkabail et al., 2006; Biggs, 2006): approximations through regressions, high resolution imaging (HRI) and georeferencing or comparing parts of the land. Gallego et al. (1993) and Gonzalez et al. (1997) used satellite images to evaluate the cultivated areas based on a quoted result from the sample lands. DeFries et al. (1996) attributed the percentage of forest land using the regressions technique. Quarmby et al. (1992) used the linear mixing method to get an estimate of region of 2500 km 2. In the agricultural environment, several uses have been implemented, for example, in the forest environment, the quantification of the wood volume (Blaschke, 2010). Likewise, the creation of new maps has helped understand the relationship between the growth of trees and the factors for its optimal agro-environmental implications. Several ways to estimate the productivity of water have been developed using satellite images (Platanov et al., 2008) by estimating the energy balance in monitoring rain (Black et al., 2016), etc. The potential development of crops is strongly related with the contribution of time and water volume; it is a fact that when an efficient water conduction network is kept, the probabilities to obtain better quality and quantity increases. The goal of this work is to present an offer to supervise the hydro agricultural infrastructure based in the acquisition of data, images and georeferencing, using the UAS with the ending purpose of getting a better idea of the functional characteristics of the hydro agricultural infrastructure, and by doing so, contribute to the crops production. 2. MATERIAL AND METHODS The procedure can be broken down in stages; in the first stage, the flight for the gathering of data begins, to be able to generate a digital model of the surface with the 2

3 Unmanned Aerial Sistems (UAS) technology. Then comes the identification of the infrastructure based on the resolution of the geometrical objects or size of the pixel (Thenkabailet al., 2007). The equipment available consists of the propeller and fixed wing (Figure 1 & 2) with their respective adjuncts for mounting the sensors. For the supervision development, twouaswhere used, one for vertical deployment and one for vertical landing following the hex copter model (Figure 1) which can be equipped with visible rank sensors and a multi-spectral RGB. Figure 1. The 6- propeller(hexacopter) with a multi spectral cámeratetracam and Visible rank. The second is a long-range SPADE 70 with independent flight (Figure 2) with a video camera and modified NDVI. With the information gathered in each flight, through image treatment using the PIX4D program, it is intended to use techniques such as the medium size of a pixel and area definitions for: A) Detection of the physical status of the hydro agricultural infrastructure; B) Visual surveillance of the irrigation systems and C) Control of irrigated areas with crops Figure 2 UAS Spade 70 with NDVI camera The first flight takes place with the visual and multispectral cameras (Figures 1 B and C). These sensors can provide visual information of the functional characteristics of the system which may be deduced through the vegetation trace.the SPADE 70 (Figure 2) is used for long-range, with independent flight for over 40 minutes, equipped with a NDVI modified camera and video. In this project, the methodology can be describe in the follow activities: 1. Identification of the area or irrigation system to flying 2. Mision planning in the especifc irrigation system 3. Mission ejecution 3

4 4. Post processing of information 5. Results presentation 2. ANALYSIS OF THE RESULTS Gomez-Candon et al., 2013 described an Object Based Image Analysis (OBIA) which presented more precise images based on pixels, especially when involved with spectral information, as it is the case here. An example is shown on Figure 3A where a body of water is represented with irrigation purposes. Through this image is possible to determine interest factors such as the start and end of the watercourse. The adjustment of this image allows the determination of the flooded area approximately of 5.5 hectares. If we add a measure to the approximation, we would be possible to have an estimate of the volume of stored water with planning purposes. Analyzing Figure 3B, in spectral format, a body of water is appreciated, which provides information with identified zones with vegetation.spectral information is very useful, for instance in the determination and monitoring of the evapotranspiration and water necessity for the crops (Xiao, 2002; Vanino et al., 2015; Saadi et al., 2015) Figure 3. Digital model of a measured area in a mariquita dam The use of spectral images allows for more rigidity in the data analysis process since there are areas where proper physical observation is not possible, for instance, in the case of a leak or micro-sprinklers (Figure 3A, B), only a spectral image will act as an indicator to evaluate performance rating of the system, through which is possible to infer if the irrigation is efficient by the uniform development of the crops (Figure 3 D), with respect to other zones and other additional parameters such as the shape and size of the plants or texture parameters of the objects shown in the image (Castillejo- Gonzalez et al., 2009; Peña-Barragan et al. 2011). On the other hand, the main problem in a irrigation systems is associate with a slow uniformity of the water distribution. This uniformity is calculated to evaluate the uniformity of distribution coefficient (UDC) and coefficient of variation (CV) of a familiar set of irrigation, The irrigation and fertigation are determined by two methods the Keller&Karmelli, And UDC = q 25 q med x100 4

5 CV = s q med x100 In sprinkler or drip irrigation this is the methology to determinate the efficience of the irrigation system (Figure 4A, B, C and D). In the case of the surface irrigation Sistem (Figure 4E), The purpose of the cutback method is to reduce surface runoff and createmore uniform infiltration.in this case the most important parameter is the time and position of the water advanced, to determinate the water simulation flow over the surface. In furrow irrigation, theshape and spacing of the furrow are limited-choice field parameters, but other parameters, such as the gradient in the downstream direction, because the existing natural slope of the land may suggest a field slope. Furrow spacing andfurrw shape ara parameters that mainly dependent on the available farm and agronomic requirements. A B SprinkleIrrrigation DripIrrigation C D PivotIrrigation SurfaceIrrigation E Figure 4. Several irrigation systems supervised with UAS CONCLUSIONS The inspections of hydraulic work with the purpose of irrigation in crops through UAS provide with the necessary information for a Type 1 inspection, according to the normative of the Mexican Water National Commission (CONAGUA) with the advantages of reducing the time and economic costs, as well as reducing risk in areas with difficult access. The use of UAS to transport sensors in one same flight 5

6 has the ability to generate information (visual, thermic, spectral, etc.) through a post process, increasing the certainty to issue reason concerning the functionality of the hydro agricultural systems; the practicality of this tool is based on the time cost and gathering of data. The geometrical features in irrigation agriculture, such as watercourses, ducts, tubes, etc. provide with useful information concerning the physical status of the hydro agricultural infrastructure. A continued supervision of this storage source, lines of conduction, ducts or tubes and irrigations systems, allows the evaluation of the physical and functional status in which they are found, and as a consequence will help estimate or quote possible losses. REFERENCES Atkinson, P.M.; Cutler, M.E.J.; Lewis, H. Mapping sub-pixel proportional land cover with AVHRR imagery. International Journal of Remote Sensing 1997, 8(4), Biggs, T.; Thenkabail, P.S.; Krishna, M.; Gangadhara R.P.; Turral, H. Vegetation phenology and irrigated area mapping using combined MODIS time-series, ground surveys, and agricultural census data in Krishna River Basin, India. International Journal of RemoteSensing2006, 27(19), Black E., Tarnavsky E., Maidment R., Greatrex H., Mookerjee A., Quaife T. and Brown M., The Use of Remotely Sensed Rainfall for Managing Drought Risk: A Case Study ofweather Index Insurance in Zambia..Remote Sens. 2016, 8, 342; doi: /rs Blaschke, T Object based image analysis for remote sensing.isprs Journal of Photogrammetry and Remote Sensing. 65; 2-16 Castillejo-Gonzalez, I.L. Lopez -Granados, F., Garcia-Ferrer, A., Peña Barragan, J.M., Jurado- Exposito, M., Sanchez De la Orden, M. and Gonzalez-audicana, M Object and pixel based analysis for mapping crops and their agro-envinronmental associated measures usiing Quick Bird Imagenery. Computers and electrnics in Agriculture 68: Conagua (2010). Comisión Nacional del Agua (2010) Estadísticas del Agua en México. México, Secretaría de Medio Ambiente y Recursos Naturales, 249 p. DeFries, R.; Hansen, M.; Steininger, M.; Dubayah, R.; Sohlberg, R.; Townshend, J. Subpixelforestcover in Central AfricafromMultisensor, multitemporal data. Remote Sensing of Environment 1997, 60, Foody, G.M.; Lucas, R.M.; Curran, P.J.; Honzak, M. Mapping tropical forest fractional cover from coarse spatial resolution remote sensing imagery. Plant Ecology 1997, 131(2), Gallego, F.J.; Delince, J.; Rueda, C. Crop area estimates through remote sensing: stability of the regression correction. International Journal of Remote Sensing 1993, 14, Gomez-Candon, D., De Castro-megias, A.I. and Lopez-granados, F Assesing the acuracy of mosaic from unmanned aerial vehicle (UAV) imagery for precisiónagricultura purposes. Precision Agriculture. DOI: /s Gonzalez-Alonso, F.; Cuevas, J.M.; Arbiol, R.; Baulies, X. Remote sensing and agricultural statistics: crop area estimation in north-eastern Spain through diachronic Landsat TM and ground sample data. International Journal of Remote Sensing 1997, 18(2), Keller J., and D. Karmeli Trickle irrigation design parameters. Transactions of the American Society of Agricultural Engineers, 17(4): Peña -Barragan, J.M., Ngugi, M.K., Plant, R.E., Six, J Objec-Based crop identification usin multiple vegetation indices, textural features and crop phenology. Remotesensingand environment. 115: PNH(2008) Programa nacional hídrico México, Comisión Nacional del Agua, 2008, 158 p. 6

7 Platonov A., Thenkabail P. S., Biradar C. M., Cai X., Gumma M., Dheeravath V., Cohen Y., Alchanatis V5, Goldshlager N., Ben-Dor E., Vithanage J., Manthrithilake H., Kendjabaev S. and Isaev S Water Productivity Mapping (WPM) Using Landsat ETM+ Data for the Irrigated Croplands of the Syrdarya River Basin in Central Asia. Sensors 2008, 8, ; DOI: /s Quarmby, N.A. Towards continental scale crop area estimation. International Journal of Remote Sensing 1992, 13(5), RemoteSens. 2015, 7, ; doi: /rs Saadi, Sameh, Simonneaux V., Boulet G, Raimbault B.,Mougenot B,Fanise P., Ayari H., andzohra Lili-Chabaane Monitoring Irrigation Consumption Using High Resolution NDVI Image Time Series: Calibration and Validation in the Kairouan Plain (Tunisia) Thenkabail, P.S.; Biradar, C.M.; Turral, H.; Noojipady, P.; Li, Y.J.; Vithanage, J.; Dheeravath, V.; Velpuri, M.; Schull, M.; Cai, X.L.; Dutta, R. An Irrigated Area Map of the World (1999) derived from Remote Sensing; Research Report No. 105, International Water Management Institute: Battaramulla, Sri Lanka, 2006; pp. 74. Thenkabail, P.S., Chandrashekhar M., Praveen N. Xueliang C., Venkateswarlu D., Yuanjie L., Manohar V., MuraliKrishna G. and Suraj P Sub-pixel Area Calculation Methods for Estimating Irrigated Areas. Sensors 2007, 7, VaninoSilvia, Pulighe G., Nino P., De Michele C., Falanga Bolognesi S. and Guido D Urso Estimation of Evapotranspiration and Crop Coefficients of Tendone Vineyards Using Multi-Sensor Remote Sensing Data in a Mediterranean Environment. Remote Sens.2015, 7, ; doi: /rs Xiao, X.; Boles, S.; Frolking, S.; Salas, W.; Moore, B.; Li, C.; He, L.; Zhao, R. Landscapescale characterization of cropland in China using VEGETATION sensor data and Landsat TM imagery. International J ournal of Remote Sensing 2002, 23,