Needs of agricultural production on the real-time monitoring, comprehensive analysis, using remote sensing technique can provide reliable information support for farmland to guide farmland management, realize the precise detection of crop condition, the field of fine management, rational use of pesticide chemical fertilizer to improve the quality of agricultural products, to break the traditional agricultural production is in a state of passive position for a long time. Modern remote sensing technology for agricultural situation is to use various active and passive detectors, satellites and various aerospace equipment without direct contact with the detection target to timely and comprehensively obtain ground object information for monitoring and analysis. Low-altitude remote sensing technology can use multiple platforms with different sensors to obtain information of large-scale farmland to help farmers timely grasp the information of crop growth and agricultural disasters, thus changing the long-term passive situation of traditional agricultural production.

Agricultural production in China is facing a less arable land resources, population explosion and the security of supply of agricultural products, and many other problems, to solve the problems of numerous factors, change the traditional agricultural production of fertilizers, excessive use of rugged phenomenon, to help farmers know reliable farmland information in time, comprehensive, many kinds of understanding of the field of crop growth status, for field management to do on time, on demand accurate job management, to save the cost of production at the same time can better protect the environment, truly make the traditional agriculture to the modernization and informatization of agriculture development. Low altitude uav remote sensing technology is not troubled by the traditional monitoring equipment's small monitoring range, low resolution, narrow field of vision and other problems, but also makes up for the satellite remote sensing by weather and cloud impact, long revisit period caused by time discontinuity and other problems. The application of uav remote sensing technology to agricultural production can timely and accurately detect agricultural situation information, and achieve positioning and quantitative management of farmland management, which can not only reduce the use of pesticides and fertilizers, but also solve problems such as less arable land area and high environmental pressure in China while improving the utilization rate. In addition, the uav can be mounted to the airborne equipment such as spray system, broadcast system, replace the traditional mechanical and artificial ground to ground crop farming activities from the air, in order to improve the work efficiency, reduce the homework of casualties caused by pesticide in the process of physical injury and ground operations mechanical damage to crops, save production cost and can improve production. However, the application of uav to guiding field management in agricultural production is still in its infancy, and many problems need to be further studied and solved. This paper systematically analyzes and summarizes the application status of uav in agricultural field from the aspects of airborne equipment, remote sensing monitoring of agricultural situation and agricultural operation, and puts forward corresponding development Suggestions for the later research in this field, expecting to be helpful for the research in this field.

1 overview of agricultural uav

Unmanned Aerial Vehicle (UAV) is an Unmanned aircraft that relies on radio communication and can fly autonomously. It is composed of flight control system, power system, radio control system, payload system and other parts. Different flying platforms can carry different payloads for various assigned missions. Uavs in agricultural production can be divided into two types: remote sensing monitoring of agricultural information and agricultural operation.

1.1 uav flight platform

According to different configurations, the uav flight platforms used in agriculture can be divided into three main platforms: unmanned helicopter, multi-rotor uav and fixed-wing uav. Other flight platforms also include ornithopter uav and unmanned spacecraft, but they are rarely used in agricultural production. Fixed-wing uav is the most ideal low-altitude remote sensing platform, because of its fast flight speed and long endurance time, it can patrol large areas of farmland to detect agricultural conditions. However, the load is small. In agriculture, fixed-wing uas are mostly used to carry small sensors for large area aerial mapping for crop growth analysis, crop yield estimation and ground object classification and identification. Unmanned helicopter more than oil and electric type, can be in situ vertical take-off and landing, and the advantages of fixed-point hovering and load is bigger, bigger rotor generates a single vertical downward flow, can penetrate the growing of crops, has the good effect, applying pesticide spraying in the agricultural production is mainly used in plant protection, can be used for spraying powder and water, but the complex operation, price is relatively high; Multi-rotor (multi-axis) unmanned aerial vehicle (uav) is the most widely used in agricultural applications. It has the advantages of reliable and stable flight, controllable altitude and speed, flexible takeoff and landing, convenient transit in plant protection operation, simple operation and low price. Multi-rotor uavs can be used in soil analysis, crop monitoring and other farmland information acquisition, pesticide spraying and other agricultural operations. In a word, the selection of the appropriate uav flight platform should take into account the price, payload, endurance time, mission characteristics and other factors, so that different equipment and uav flight platform can be properly matched to achieve efficient operation effect.

1.2 airborne earth observation sensor

According to China's traditional agricultural production experience, the environment of farmland planting operations greatly affects the quality and yield of crops. Therefore, how to obtain more accurate planting environment to understand farmland information has become the key research direction of China's agricultural development towards informatization and precision agriculture. At present, the information collection of small farmland in China mainly adopts monitoring stations, manual detection and other methods. The collection efficiency and accuracy are limited to meeting the needs of small farmland. However, it cannot meet the needs of large-scale farmland environment monitoring, and the uav can better adapt to the monitoring of large areas of farmland and the effective acquisition of information by carrying different earth observation sensors. According to different requirements of farmland environment monitoring, uavs can carry digital cameras, multispectral cameras, hyperspectral cameras and lidar sensors for efficient and all-ground information collection, so as to help farmers and the masses understand farmland information and better guide agricultural production based on this. According to the difference in the range of the monitored bands, it can be divided into digital cameras in the visible band of 400 ~ 760 nm, multi-spectral cameras in the visible near-infrared band of 400 ~ 1100 nm, and thermal infrared band cameras in the range of 3.6 ~ 13.5m. Optical digital cameras are widely used because of their low cost and simple operation. Aerial images can obtain gray scale or color images of blue (450-520nm), green (520-600nm) and red (630-690nm) bands in crops. They can intuitively obtain phenotypic characteristics of various crops and invert various parameters, mainly used in crop classification, insurance claims and other fields. The price of the multispectral camera is moderate, which can obtain the multi-spectral information, texture information and structure information of the ground object in blue, green, red and near infrared. It can extract a variety of parameters used in the fields of crop growth, soil water stress, ground object identification, fine classification, vegetation coverage index extraction, leaf area index extraction, pest monitoring, yield estimation, etc. Hyperspectral cameras have high cost, strong continuity of image spectral bands, and large computational cost. Obtaining more detailed spectral information of crop canopy is conducive to more accurate inversion of ecological and physiological parameters of ground objects, which is mainly used for inversion of physical and chemical parameters such as chlorophyll content and water content of leaves in fields such as crop growth potential and soil water stress. The thermal infrared camera can obtain the information of crop canopy temperature to analyze the effects of environmental stress on leaf photosynthesis and evapotranspiration rate. With high cost and large amount of image calculation, lidar sensor can quickly and effectively obtain surface point cloud information for inversion of plant height and biomass of crops. In general, remote sensing monitoring of agricultural situation with large scope, high timeliness and objective accuracy needs to be equipped with different types of sensors, and the rapid and lossless acquisition of agricultural situation information requires comprehensive consideration of the characteristics, use, cost and other factors of sensors.

1.3 airborne agricultural machinery equipment

Unmanned aerial vehicles can carry spraying system, seeding system and other human resources to carry out field operations, which can effectively solve the deficiencies in efficiency, quality, labor intensity and safety of manual operations. Drones can carry a variety of high, intelligence information of remote sensing monitoring equipment to get accurate and comprehensive information of farmland, and coupled with empty unmanned aerial vehicle (uav) build farming operation integration of heaven and earth of farmland management system, and achieve high yield, high efficiency and low cost while reducing the use of pesticides and water, reduce the pollution to soil and environment. In terms of uav pesticide spraying and seed seeding, liu zefeng et al. designed the variable spraying control of the liquid pump of the uav spraying system based on Pulse Width Modulation (PWM) technology, which reduced the overuse of pesticides. The seeding uav produced by zhuhai yuren agricultural aviation co., LTD has completed the precision aerial seeding of rice in shantou city, which has improved the mechanization level of rice sowing. Zhang Houdong to unmanned aerial vehicle (uav) as the carrier in the xinjiang region, such as seeding hairy leaves Chinese trumpet creeper, unmanned aerial vehicle (uav) by not limited by topography and geomorphology, high efficiency, save manpower, reduce labor intensity greatly. At present, there is still a lot of research work to be done on the precision spraying of uavs based on the severity of pests and diseases in farmland. In terms of solid fertilizer and fodder, there are some problems such as uneven distribution of particles along with airflow. There are some limitations in the aerial seeding of uav. The preliminary preparation work is relatively long, and the distribution uniformity is greatly affected by the seed size and density.

2. Application status of agricultural uav

2.1 application of uav remote sensing in agriculture

2.1.1 crop growth monitoring

The main purpose of crop growth monitoring is to monitor and analyze crop growth status and trend. Pei haojie et al. used the hyperspectral images of crops obtained by unmanned aerial vehicles to screen the spectral indexes with good correlation with crop growth from NDSI, RSI and SSI, and the established linear regression model could well monitor the growth status of wheat. Wang chulfeng et al. found that there was a high correlation between the normalized difference index (NDI) calculated from the image of visible band obtained by the consumer camera mounted on the uav and the measured normalized difference vegetation index (NDVI) on the ground, which could be used to accurately obtain crop growth information for remote sensing monitoring of rape seedling growth. Bala et al. used Terra Modis reflectance data to estimate potato yield in monsiganj district, Bangladesh, and found that vegetation index was an effective tool for early potato yield estimation. Brovkina et al. successfully applied the analysis based on NDVI and PDC gratings to the separation of spruce and silver fir by using uav remote sensing monitoring technology, and identified the category of dead trees, proving the potential of NDVI's qualitative classification of spruce trees. This method can be applied to monitor the health status of spruce in local forest areas. At present, multiple linear regression and least square modeling are widely used in the monitoring of growth and yield estimation of target crops. The spectral information combination of different bands obtained by the uav with different sensors is used to form the vegetation index. Further research is needed to establish the growth monitoring model of specific crops at different growth periods.

2.1.2 crop yield estimation by remote sensing

Crop yield estimation by remote sensing can realize accurate estimation of plot area and yield, which is of great significance for large-scale agricultural operation and management. The remote sensing production estimation system composed of unmanned aerial vehicle (uav) sensors is characterized by fast information acquisition speed, low cost, flexibility and high efficiency. Therefore, the optimal vegetation index can be obtained during the optimal operation period to establish an effective uav remote sensing production estimation model. Leon et al. used a digital camera mounted on a uav to quickly and losslessly obtain canopy image information of rice from heading stage to maturity stage, and proved that the use of digital images and k-means clustering algorithm could accurately estimate rice yield. Gong et al. made remote estimation of rapeseed yield by using a micro-digital camera mounted on uav combined with spectral hybrid analysis technology, and found that remote sensing of canopy spectral reflectance (VI) of uav was the most accurate method for estimating rapeseed yield. Somard et al. used the technology of combining RGB images obtained by drones with ground information to use object-based image analysis (OBIA) to estimate the sugarcane yield in Thailand quickly and reliably. Zhao et al. used the Cuber UHD185 imaging spectrometer mounted on the uav to obtain stable soybean canopy spectral data and accurately estimated the soybean yield in jixiang county, shandong province. At present, the model of crop yield estimation based on crop canopy image data needs to be further verified and optimized, so as to provide a fast and efficient low-altitude management scheme for large-scale agricultural planting and operation, guide reasonable arrangement of agricultural production, and accelerate the implementation of policies such as agricultural subsidies, verification and declaration.

2.1.3 diagnosis of crop nitrogen nutrition

Nitrogen is an indispensable nutrient element in the whole growth cycle of crops and is crucial to improve photosynthesis and productivity. Crop n nutrition diagnosis plays a key role in improving crop yield, quality and soil fertility, and can provide decision support for agro-technical departments and agricultural production managers to achieve sustainable agricultural production. Unmanned aerial vehicles (uavs) are equipped with near-earth sensors to obtain spectral information of crop canopy, establish nitrogen diagnosis model, optimize geographic information system and expert decision system, and apply precise variable fertilization widely at home and abroad. Liu changhua et al. used the vegetation index obtained by Tetracam mini-mca 6 multispectral camera on a uav to indirectly diagnose the nitrogen nutrition of crops. Nasi et al. estimated the biomass and nitrogen content of barley and grass by using the three-dimensional characteristics of uav and airplane-based spectroscopy and photogrammetry, and the results showed that the results obtained by the hyperspectral camera in the nitrogen content estimation were relatively accurate. Zheng et al. combined uav-based multispectral images with ground hyperspectral data to estimate rice plant nitrogen concentration (PNC) by combining selected vegetation index (VIs) and texture information. Liu et al. conducted quantitative modeling of nitrogen content in winter wheat leaves by using the hyperspectral data based on uav, and found that the BP neural network modeling results were generally better than the multiple linear regression modeling results, which could accurately estimate the nitrogen content in winter wheat leaves at jointing, leaf and flowering stages. The diagnosis of crop nitrogen nutrition is particularly important in the field of accurate fertilization. Accurate evaluation of crop nitrogen content and rational optimization of field fertilization can not only achieve high quality and excellent crop yield but also reduce environmental pollution caused by excessive fertilization. However, at present, crop nutrition diagnosis based on uav digital image still remains in the establishment of crop diagnosis model, and further research is needed on how to apply the formula diagram generated by nutrition diagnosis to guide variable fertilization in agricultural production.

2.1.4 crop pest monitoring

Crop diseases and insect pests disaster is one of the main natural disasters, the large-scale epidemic, fulminant, destructive diseases and insect pests cause a large number of food losses every year, seriously affecting the quality and yield of agricultural products. Annual losses from disease and pests account for more than 14 percent and 10 percent of world food production, respectively. In China, the losses caused by diseases and insect pests account for about 10%~15% of all kinds of agricultural disasters. The outbreak of crop diseases and insect pests from a small range to a large scale epidemic or destruction is closely related to the climate characteristics, crop varieties, planting habits and control situation of the planting area. It is of great significance to suppress the outbreak of disease and insect pests and reduce the loss of grain output by using uav's characteristics of low-altitude remote sensing, real-time and flexible, so as to timely monitor the location of disease and insect pests, achieve early detection and carry out early effective prevention and control. At the same time, it can effectively analyze the outbreak mode and spread characteristics of different diseases and insect pests, and establish the evaluation model and diagnosis model of different diseases and insect pests for different crops, which is of great significance for rapid and effective control. Aylo et al. used drones and Lagrange models to track the distribution of potato late blight bacteria in the atmosphere, to help predict the risk of disease transmission between adjacent potato fields, and to provide effective assistance for the entire regional decision support system. Hunt et al. used the uav to carry a multi-spectral camera and found that the image analysis based on specific targets could accurately estimate the extent of potato beetle pests. Stanton et al. used a small fixed-wing system with a near-infrared camera to evaluate the damage caused by aphid invasion of sorghum by the negative correlation between aphid density and NDVI in the test area. Huang wenjiang et al. proposed a new solution to the detection of crop diseases and pests in large areas of farmland. To sum up, at present, remote sensing inversion models for crop diseases and pests monitoring based on uav remote sensing still need to be built for different crops and different pest levels, so as to accurately monitor the outbreak location and damage degree of pests, accurately and timely prevent and control, and reduce the losses caused by pests.

2.2 application of uav agricultural operations in agriculture

2.2.1 application of uav in aerial drug application

With the speeding up of urbanization construction in recent years, a large number of rural labor to urban rural population aging phenomenon is serious, the human cost, together with the emergence of new large-scale operational farm and a large area of outbreak of diseases and insect pests on efficiency and precision applying pesticide, puts forward new requirements, create a good condition for agricultural drone aircraft operations. Agricultural plant protection uav has outstanding technical advantages and has the following advantages:

1) high efficiency. The scale of operations can be up to 8~11 hm2/h(8 hm2/h for forest trees), which is more than 3 times more efficient than ground machinery and 100 times more efficient than artificial spraying, which can greatly liberate labor force.

2) no terrain restrictions. The flight of agricultural uav is not restricted by complicated geographical factors such as mountains and hills, and the aerial operation avoids the damage to crops.

3) good control effect. Assignments and adopt a low-level suspense sinking airflow variable spray, good degree of liquid atomization and deposition quantity and coverage density high, high speed rotation to produce vertical downward flow can not only reduce the droplet drift and enhanced liquid penetrating, crop and foliage can be evenly by medicine, and a long time to deposit on crop leaves, is advantageous for the use of diseases and pests in order to achieve the optimum insecticidal effect.

4) low cost of plant protection. According to the calculation of a large number of practical applications, using agricultural uav to spray pesticide not only saves water and medicine but also costs much less than ground machinery and artificial spraying.

5) safety and environmental protection. Uav spraying pesticide can significantly reduce the application amount of pesticide, reduce the environmental pollution caused by pesticide and reduce the physical injury to operators, which is in line with the national requirements of reducing the application of chemical fertilizers and pesticides and increasing the efficiency.

6) emergency solutions for major diseases and pests. In the year of major diseases and pests, agricultural drones can quickly assemble and set up a large-scale service network for emergency rescue, effectively restraining the spread of disaster.

China's aviation application started relatively late, which can be traced back to the early 1950s, from the initial manned fixed-wing aircraft and manned helicopter to unmanned helicopters and multi-rotor uav. At present our country agricultural uav flight control technology is a world leading position, even unmanned aerial vehicle (uav) equipment and work area is the world's first, total ownership in the plant protection machine market has more than 30, 000, operating area MuCi growth in 2017 from 106 million to 2018, more than 300 million MuCi, marks the development of China's aviation plant protection into the new era.

Japan began developing spray drones in the 1980s and was the first country to use unmanned helicopters for agricultural production. Developed in Japan in 1987 will be the first unmanned helicopter R50 used for pesticide spraying, eppo unmanned aerial vehicle (uav) after nearly 30 years of development has experienced from scratch, from some helicopters to the rapid development of the unmanned helicopter, YAMAHA company research and development of unmanned helicopter eppo successfully solved the Japanese restricted by the terrain and the land area is not suitable for manned aircraft assignment problem, making an important contribution to the development of Japanese agricultural aviation. In 2003, for the first time, unmanned helicopters were able to control more rice than manned helicopters and the rate of operation has been higher than manned helicopters ever since. From 1995 to 2015, the number of uas in Japan increased from 307 to 2,668. By the end of 2010, Japan had used agricultural uavs to control 963,000 hm2 of farmland, accounting for 38% of the total area of aviation operations, with 14,163 registered uavs. American agricultural aviation began in 1906 and has developed for more than 100 years. It has a sound aviation organization system and quality assurance model. The United States is the country with the most extensive application of agricultural aviation. There are more than 2,000 agricultural airlines, about 4,000 agricultural aircraft in use, and more than 3,000 registered and licensed pilots. The cultivated land area of the United States is nearly 34 million hm2 per year, accounting for more than 40% of the annual agricultural aviation area of the United States. In the United States, forest flight prevention and control and rice flight prevention and control of the use of aviation flight prevention operations.

2.2.2 application of uav in field management

In field management, unmanned aerial vehicle (uav) in addition to can be used for spraying pesticide can also be used for crop planting, pollination, fertilization and so on, from the air instead of manual operation reduces the operation in the process of agricultural machinery of the damage to the crops and soil recompacted, high working efficiency and quality and reduce the labor intensity of the ground operation staff, to achieve large area, large-scale production. During maize pollination, kong deshen et al. used aerial pollination assisted by plant protection uav, and found that two times of assisted pollination increased the yield of maize mu by 97.10kg and reduced the baldness rate by 7.5%, while one time of assisted pollination increased the yield of maize mu by 58.67kg and reduced the baldness rate by 6.13%. Li jiyu et al. studied the effect of rotor wind field on the distribution of rice pollen and provided relevant references for the aerial pollination of uav. In terms of unmanned aerial seeding, China has applied the technology of unmanned aerial seeding to rice seeding. During the spring tillage in 2019, it carried out unmanned rice seeding for the majority of farmers in bayi village, yashao town, yangjiang city, guangdong province, successfully solved the problem of low efficiency and high cost of manual rice transplanting. Gao zhizheng et al. discussed the characteristics, application direction and current shortcomings of the broadcasting technology based on the application of uav broadcasting technology in agriculture. The technology of uav broadcasting is feasible and superior in agricultural production management, and it is an important link to accelerate the modernization of agricultural production management in China.

3 outlook

In a word, the application of uav remote sensing monitoring and agricultural operation in agricultural production is of great significance for improving the efficiency and quality of agricultural production and promoting the development of smart agriculture and precision agriculture in China. Uavs sensor of remote sensing monitoring of timely access to accurate information, farmland on crop growing and soil information, such as agricultural disasters effectively dynamic monitoring, and satellite remote sensing monitoring and ground sensors to monitor complementary and build integrated monitoring system of heaven and earth, generate accurate farmland "diagnosis" and combining the actual situation of farmland used to guide production field, field management do it on time according to need to precise fertilization, administer and crop irrigation management.

At present, there are still the following problems in the application of uav in remote sensing monitoring and agricultural operations. Uav monitoring data are coupled with ground monitoring and agricultural machinery data. In terms of agricultural operations, the service life of key components of uav cannot fully meet the application needs; Lack of aviation plant protection agents and auxilaries for the use of uav; Lack of a standard system for uav industry applications. In order to promote the better application of agricultural uav in agricultural production, the following aspects of research will be the focus.

3.1 combination of uav technology and precision agriculture

Precision agriculture is the combination of 3S(GPS, GIS and RS) and modern agriculture in the agricultural production process of agricultural resources and agricultural operations timing, positioning, quantitative scientific management, with minimum input to achieve high-quality and high-yield sustainable development of agriculture. With its flexible and efficient characteristics, uav can provide accurate information support for all aspects of agricultural production, and provide reliable basis for fertilization, application and irrigation for farmland production managers. Uavs sensor for farmland soil moisture, vegetation index, crop nutrient stress monitoring information model of inversion for different crops, the farmland is divided into many small operating unit for management, the accurate, according to different position of crop growth in farmland irrigation, spray, and partition of variable rate fertilization in order to achieve efficient allocation of productive factors, such as unmanned aerial vehicle (uav) can according to different location of farmland under different degree of insect pest of crops on variables of nozzle precision pesticide applying intelligent control, unmanned aerial vehicle (uav) technology combined with precision agriculture development can greatly promote the modernization process of precision agriculture.

3.2 uav is used in conjunction with aerial plant protection agent

It is well known that in the process of spraying pesticide by uav, the droplet of traditional Chinese medicine is subject to the combined action of rotor wind field, head-on wind force and environmental wind force. At present, the supporting special preparation of plant protection uav is relatively lacking or the technology is not mature enough, so the plant protection can only be carried out by increasing the concentration of conventional pesticides. At present in China pesticide information online is used with eppo unmanned aerial vehicle (uav) special agents or additives registration, many pesticide manufacturers and drones leading enterprises began to fly against the development of agents and auxiliaries, by changing the droplet diameter size and evaporating inhibitor and settling agent is used to reduce the degree of droplets falling in the process of evaporation. Pesticides in the department of agriculture set to 2020 action plan for zero growth in regulation to reduce pesticide use, only with nano pesticide intellectual property rights developed by nanjing, good at thinking of biological science and technology co., LTD. Nanometer spray pesticide in more using unmanned aerial vehicle (uav) for prevention and control of pest control experiment and obtained satisfactory results, unmanned aerial vehicle (uav) nanoparticle preparation once into widespread use, can reduce more than 30% of the existing pesticide use in our country, give full play to its high efficiency, safety, economy and environment friendly characteristics, will greatly promote China's pesticide reduction efficiency.

3.3 improve relevant system standards

Due to the low threshold of China's uav market, the lack of regulatory laws and regulations, and the lack of sound relevant industry standards, the national subsidy for the purchase of uav agricultural machinery is only implemented in some areas, but not widely popularized and applied nationwide. As constraint agricultural unmanned aerial vehicle (uav) flight and ensure the healthy and orderly development of industry, the safety of the relevant departments should break through the region limit at the national level to improve the quality of unmanned aerial vehicle (uav) technical requirements, operators of driving training, job specification and so on industry standard system construction, accelerate agricultural machinery purchase subsidy policy landing unmanned aerial vehicle (uav) and its application in industry development. The construction of the standard system in the fields of uav plant protection application standard, prevention effect evaluation standard and operation environment risk evaluation standard will also greatly standardize the industry application of uav plant protection.

4 conclusion

Agricultural uav can acquire agricultural information timely and accurately and assist agricultural operations from the air. It is an important means of application in the field of precision agriculture and has incomparable advantages compared with other traditional monitoring methods and traditional ground operating machinery. With the advancement of agricultural supply side structural reform in our country, the implementation of land circulation and the occurrence of large-scale operational farm agricultural unmanned aerial vehicle (uav) for production manager provides a new way of development, management after diagnosis, timing of field operation management, positioning and quantitative precision management will greatly promote the agricultural development in the direction of intelligent, modernization.