This article is produced by NetEase Smart Studio (public number smartman 163). Focus on AI and read the next big era! [Netease Smart News November 11 news] UAVs travel back and forth in a row in the head, robotic equipment on the ground busy working. Orbiting satellites can capture high-resolution images of the ground. The glimmers of light before dawn flood the earth, but they can't see the figure of a person. Of course, this is not the post-apocalyptic landscape of the Terminator, but the state of the future farm. Every aspect from seeding to harvesting is likely to be automated sometime in the future without making the human hands dirty. In fact, this kind of science fiction story has already appeared in real life. Today, artificial intelligence robots can eliminate weeds with extraordinary precision, and automatic tractors continue to work continuously in farmland with high efficiency. Satellites can assess the health status of crops from locations far away in outer space and provide large amounts of data for business intelligence that was once available only to Fortune 500 companies. “Precision agriculture is about to enter a new stage of development. At this stage, smart machines can operate on their own, which can greatly increase the efficiency of agricultural production. That is, precision agriculture is about to develop into robotic agriculture,†said a robotic agriculture in Asia sometime last year. At the conference on development status, Professor Simon Blackmore talked about this. Blackmore is director of engineering at Harper Adams University and head of the National Center for Precision Agriculture in the UK. Recently, the university where Blackmore is located looks forward to the future. This project, known as “no-tillage hectares,†was led by researchers of Hapa Adams and private companies who planted spring barley on a hectare (approximately 2.5 acres), without the expense of any manpower. The team redesigned, combined and robotized agricultural equipment. The agricultural machinery involved included Japanese tractors and combine harvesters that had been in use for 15 years. In addition, drones act as scouts, investigate operational procedures and collect samples to help the team monitor the production of spring barley. At the end of the season, "robot farmers" harvested about 4.5 tons of spring barley worth 200,000 pounds. “This project is to prove that from a technical point of view, humans do not personally work, the farm can also achieve the harvest. And we have already achieved this.†Precision Decisions researcher Martin Abell at a press conference Say this. The company has worked with Harper Adams. Robot farmer As a recent example, Harper Adams's experiment shows how the machine affected the agricultural industry. While the unmanned farm harvester harvested barley, John Deere announced that it will acquire a startup named Blue River Technology for $305 million. Blue River has developed a system called "see-and-spray" that combines computer vision and artificial intelligence to distinguish between crops and weeds. The system can automatically identify chemical fertilizers for crops and use weeders to remove weeds. The system has a very high degree of accuracy and can eliminate 90% of the chemicals needed in traditional agriculture. The industrial sector that benefits from the development of robots is not just agricultural farmland. A California company called Abundant Robotics, formerly known as SRI International, a non-profit research organization, is now developing a robot that can pick apples with a vacuum arm that can suck fruit directly from the orchard's tree. “The traditional robot design philosophy is to have the robot perform a specific task repeatedly. However, the robots used in the food and agriculture industries must be more flexible than what we see in automobile manufacturing plants. Therefore, they can better respond to food. Natural changes in the production or outdoor environment,†said Dan Harburg. He is a partner in Anterra Capital, a venture capital firm. He previously worked for a start-up company in Massachusetts, whose business includes the manufacture of robotic arms that can pick up fruit. He pointed out: "This means that startups making alternating gradient focusing robots must design systems from scratch, and spend a lot of time and money, and their robots must be able to accomplish multiple tasks in order to avoid making robots for most of the time. Are idle on the shelves." Altitude detection To successfully grow some crops, not only to produce a large number of robotic tractors. The future of the farm will rely on drones, satellites and other onboard instruments to provide data on ground crops. For example, companies like Descartes Lab use machine learning techniques to analyze satellite imagery to predict soybean and corn production. Los Alamos, a Mexican startup, collects 5 trillion bytes of data from several satellite constellations including NASA and European Space Agency on a daily basis. In conjunction with weather data and other real-time input signals, Descartes Labs can predict corn field production with 99% accuracy. Its artificial intelligence platform can even assess the health status of crops through infrared detection. The US Department of Defense Research and Planning Agency recently allocated $1.5 million to the Descartes Laboratory for monitoring and analysis of wheat production in the Middle East and Africa. They believe that accurate forecasting results may help identify crop failure areas, thereby effectively avoiding famine and political turmoil. Another company, TellusLabs, was originally based in Somerville, Massachusetts. They use machine learning algorithms to predict corn and soybean production with comparable accuracy to satellite images. At the same time, farmers do not need to enter the farmland themselves to understand the situation. Ceres Imaging, a startup based in Oakland, can produce high-resolution images using a miniature airborne multispectral camera that shuttles through the field. These snapshots can capture landscapes at different wavelengths, reveal problems such as water pressure, and provide estimates of chlorophyll and nitrogen levels. Using geo-tagged images can help farmers easily locate problem points. Increase internal space Even the best intelligent systems—whether from drones, satellites, or machine learning algorithms—are faced with huge challenges in predicting unknown problems brought about by climate change. This is one of the reasons why more and more companies are investing in so-called "controlled environmental agriculture" on the farm. Today, this means more than just configuring an expensive greenhouse, but also a warehouse-sized, fully-automated vertical farm to provide more space for robotic operations, and the farm is not located in Kansas or Nebraska. An empty area, but in the middle of the main street in the United States. Those who support this new concept farm believe that these high-tech indoor farms can achieve higher yields, while drastically reducing water use and the use of synthetic materials such as fertilizers and herbicides. The "Timber Corporation" from San Francisco is developing an acre of urban greenhouses that are operated by robots and allegedly produce approximately 30 acres of farmland. Using artificial intelligence, a team of three robots can perform the entire process of planting, cultivating and harvesting crops. San Francisco-based vertical farm startup Plenty, which is also headquartered in San Francisco, is committed to using AI technology to automate operations. Earlier this year, it received a $200 million trust vote in the Softbank Vision Fund. The company claims that the system uses water as 1% of traditional agricultural water, but its output is 350 times that of traditional agriculture. Plenty products represent the latest achievements in the concept of urban farms, such as Bowery farms and aerof farms. In an interview with AgFunder News, Dr. Gary Stutte, a specialist in space agriculture and controlled environmental agriculture, said: “I can imagine that building a larger-scale indoor farm in an economically backward and food-poor area will not only stimulate macroeconomic development. , create more employment opportunities, and can increase the income of the region.†“The model of indoor agriculture can be developed into an economic growth and food security engine in the urban and rural food deprivation zone.†However, this model also faces challenges and criticism. Most of the products that these farms can produce fall into the category of “green leafy vegetables†and are often expensive, which seems to run counter to the mission of creating “oasis†in urban “food desertsâ€. Although these farms may minimize the amount of water used, the power required for operation, especially the use of LED lights, is not low (lights play an important role in radically changing indoor agriculture). However, all these technological advances, including robotic farmers and automated greenhouses, will face nearly 10 billion resident population in 2050. A data frequently cited by the Food and Agriculture Organization of the United Nations (FAO) shows that worldwide food production must increase by 70% to meet the population's demand for food. Perhaps science and technology cannot save the world, but it helps to solve the problem of human food and clothing. (Source: Singularityhub Compilation: NetEase Execution Translation Robot Review: Sarah) Pay attention to NetEase smart public number (smartman163), obtain the latest report of artificial intelligence industry.