Pinduoduo Strawberry Planting Competition: Algorithm And Data Help New Agriculture

On October 15, after nearly three months of competition, the first season of the first strawberry AI planting competition in China ended, and the artificial intelligence team was overall ahead of the top farmers' team.

On July 20, under the guidance of the food and Agriculture Organization of the United Nations, the first "Duoduo agricultural research and technology competition" jointly organized by China Agricultural University and pinduoduo was launched in Kunming, Yunnan Province. Four AI teams from the world and four top farmers from strong strawberry planting counties in China launched a 120 day plateau strawberry planting challenge.

Now, in the middle of the competition, the first season of strawberries is in high yield. The field data show that: the sugar content of strawberry of AI team and the top farmers' team are the same, but AI team leads in the overall yield, which is more than 175% higher than the top farmers' team on average.

At the same time, AI team's "algorithmic planting" has also made phased achievements, such as clustering algorithm, image recognition algorithm, collision algorithm, multi-layer neural network and so on, which is expected to explore "smart plug-in" for domestic strawberry planting and realize the optimal planting with one button.

"This is a" man-machine competition ", but it is also necessary to explore man-machine cooperation and digitize the artificial experience, so as to provide a reference model for the production area." LAN Ke, head of the Organizing Committee of Duoduo agricultural research competition, said that the current data does not represent the final result. He hopes that the AI team and the top farmer team can find out an optimal solution to help small and medium-sized farmers improve their benefits.

AI team leads in many indicators

Strive for "one click planting"

In the first production season since September, there have been differences in strawberry production between four AI teams and four top farmers.

According to the data measured by the expert team of Yunnan Academy of Agricultural Sciences and the technical support Party of the competition, the four AI teams of Zhiduo berry, AICU, njai. Berry and cyberfarmer. Hortigraph are all in the lead in comparison of fruit number in the mature cycle, and the overall average value exceeds the top farmers' team by more than 175%. On the average fruit weight index, cyberfarmer. Hortigraph, njai. Berry and Zhiduo berry were all in the leading position.

In the eyes of Dr. Ruan Jiwei, a researcher at Yunnan Academy of Agricultural Sciences, the eight teams in the competition "showed their magic powers" and quickly adapted to the strawberry planting operation under the plateau climate, and successfully advanced the strawberry production season in Yunnan by more than 60 days.

"At the beginning of planting strawberry in the middle and lower reaches of the Yangtze River, the first season of the competition has already had a good harvest. This alone can fill the gap in the consumption of fresh strawberry in summer and autumn in Yunnan." Ruan Jiwei said. In order to support the competition, the expert team of Yunnan Academy of Agricultural Sciences also provided two national technological invention patents, which were independently developed by Yunnan Academy of Agricultural Sciences.

During the two months of the competition, AI team continuously explored new technologies to help farmers realize "one click planting".

In four independent AI greenhouses, the intelligent environmental control computer and water and fertilizer integration system of Ridder group are applied. It integrates more than 50 sets of sensors, which can read 45 production environment parameters such as temperature and humidity, light intensity value and carbon dioxide concentration in real time, and automatically control more than 30 irrigation and ventilation parameters according to preset conditions.

More than 40 young scientists from different disciplines and fields will challenge the use of innovative algorithms to improve their AI models and realize "unattended" strawberry planting with artificial intelligence through the special API interface for competition.

"Through clustering algorithm, image recognition algorithm, collision algorithm and other strategies, we are exploring and training a more intelligent AI model. It's like the doctor's automatic control of the growth of strawberries. " Cheng Biao, the leader of Zhiduo berry, said that AI agriculture must not be an expensive laboratory technology. They hope to explore solutions that are really applied to Yunnan through the competition, and provide cost-effective solutions for small farmers.

Li Daoliang, a professor of agricultural science and technology in China, believes that the new generation of agricultural science and technology competition has a big difference from the traditional agricultural science and technology competition Intelligent decision-making and intelligent operation are also the core of unmanned agriculture in the future. "

Deep ploughing AI agricultural data to help development

How popular is artificial intelligence, how urgent is the demand of the times. The food and Agriculture Organization of the United Nations predicts that by 2050, the global population will exceed 9 billion, and the demand for food will increase by 70%. How to maximize production capacity and achieve sustainable development, artificial intelligence is one of the solutions.

Pinduoduo's eyes have never left agriculture. Different from the Standardized Agriculture in the United States, China's agriculture as a whole presents scattered small-scale farming. In the whole agricultural cycle, farmers are in the information blind area, and the judgment of their subjective will is facing a market full of uncertainty. In addition, the distribution and circulation structure between farmers and consumers is complex, farmers are often forced to harvest early, and consumers can not get the same value of products.

The data information of pinduoduo platform is centralized, which makes the mode of collage gather consumer demand and further transmit to the upstream farmers. The essence of this model is to enable manufacturers' data and give them more time to plan production and match consumer demand.

In 2019, pinduoduo's total orders reached 19.7 billion, and the number of active agricultural (by-product) businesses on the platform reached 586000, up 142% year-on-year.

Gong Ze, senior director of pinduoduo new agricultural and Rural Research Institute, said that although there are obvious differences in geographical environment, varieties, and mature seasons in different regions of China, the platform can quickly aggregate the scattered and long-term needs of more than 600 million users into short-term batch demands through the transformation of the consumer side.

Pinduoduo, as the largest platform of China's agricultural products, is the promoter and witness of this agricultural reform. Pinduoduo started from agriculture. On the basis of "pinduoduo", the platform gradually developed a mature "farmland cloud spelling" mode. After the demand of products is integrated in the cloud, it is linked to the major production areas through the platform of new farmers system all over the country, and it is matched to the corresponding consumers according to the variety, maturity cycle, logistics conditions and precision. This model is based on data algorithm, this competition is just a game of further algorithm and experience.

According to the financial report of pinduoduo, in the first quarter of 2020, the number of orders for agricultural products exceeded 1 billion, with a year-on-year increase of 184%. The turnover of agricultural (sideline) products in 2019 reached 136.4 billion yuan, up 109% year-on-year. The sensitivity of data makes pinduoduo see the possibility of combining artificial intelligence with agriculture.

Strawberry planting competition can be seen as a flag waving of pinduoduo's AI agriculture. "A small step forward in technological progress may solve many problems faced by small-scale farmers and bring about a more stable supply chain of agricultural products." According to gongze.

"The purpose of the competition is not simply to win or lose, but to combine the new generation of artificial intelligence technology with agricultural production through the same platform of the old generation of farmers and the new generation of technology, so as to provide a lower cost and easy to operate digital agricultural production management method for China's numerous agricultural product industrial belts and small farmers." Gongze told the 21st century economic reporter. This is another achievement driven by artificial intelligence.

?