The former is deeply involved in the agricultural field, focusing on agricultural machinery manufacturing; the latter is known as the global leader of programmable graphic processing technology. Just like this, the two companies, which seem to have no intersection, have recently started cross-border cooperation hand in hand.
Kubota self driving tractor concept car display
Appearance science fiction new trend, different from traditional agricultural machinery
According to Nikkei Asia, Kubota and NVIDIA have recently reached a cooperation to develop highly precise automatic driving agricultural tractors.
In Japan, since the early 1970s, it has stepped into an aging society. In 2010, the proportion of the population over 65 years old has increased to 23%.
With the aggravation of the aging population, there are a series of problems in the agricultural field of Japan, such as the shortage of labor force, the low efficiency of work, the increase of abandoned cultivated land, and the difficulty in promoting production smoothly.
Data show that more than half of Japanese farmers are over the age of 65
Facing increasingly severe challenges, Japan aims at agricultural mechanization and automation, and vigorously develops unmanned agriculture.
Since 2003, the Japanese government has been implementing the next generation of agricultural machinery emergency development plan, vigorously promoting 46 kinds of high-performance agricultural machinery, including automatic rice transplanter, driving vegetable cultivator, harvester, infrared dryer, belt packer, pesticide sprayer, etc.
Automatic rice transplanter launched by Japanese Agricultural Company Yanmar in 2019
These automatic agricultural machinery, let Japanese farmers taste a great sweet, they save at least half of the labor cost. The data show that the automatic transplanter can plant 10 mu of rice seedlings a day, while the manual transplanter can only complete 1 mu of work per day.
Therefore, as an old Japanese agricultural machinery manufacturer, Kubota has been actively looking for methods in recent years, hoping to realize the popularization of agricultural automation with the help of science and technology. The company said it would save labor through automated and driverless agricultural machinery, and use data to achieve precision agriculture..
Founded in 1890, Kubota is the largest agricultural machinery manufacturer in Japan
Automatic driving makes tractor cyberpunk
In recent years, Kubota has been walking on the road of tractor automatic driving.
In 2017, Kubota has launched the agrirobo series of automatic driving tractors. However, these tractors only use radar to determine the surrounding environment and realize autonomous operation. The operation process still needs manual supervision and does not use AI technology.
Agrirobo tractor can be driven automatically or manually
Next, Kubota said that through cooperation with NVIDIA, the technology it developed is expected to achieve full automatic driving.
Due to the complex working environment of tractor in farmland, the soil is loose and rough, and if you dont pay attention to it, you will make mistakes. Therefore, for automatic tractor, it is very important to master the surrounding environment timely and accurately, and to process the collected data without delay.
Kubota will equip the tractor with NVIDIA GPU, artificial intelligence module and camera to realize the function of real-time processing of road and environmental information, and fine analysis of weather information and crop conditions.
In product development, Kubota will use NVIDIA Jetson computing platform (an embedded AI platform for NVIDIA edge devices) to carry out research and development. Selecting NVIDIA Jetson platform has the following two advantages:
NVIDIA Jetson has high computing power, accuracy and energy efficiency, while the industrial version has a highly durable design, which can meet the requirements of harsh environment in agricultural machinery.
In addition, on the edge side, real-time smoothing is needed for high-definition screen splicing and edge detection, and NVIDIA Jetsons processing ability in this respect is evaluated as the best.
Kubota self driving tractor concept car, powered by lithium batteries and solar cells
It looks like a mecha in appearance
The DGX system can analyze the input information from many cameras installed on agricultural machinery, repeatedly train the AI model, and return the results to NVIDIA Jetson to improve the accuracy of the model.
Kubota is focusing on the next generation of agriculture. In the field of agriculture, we comply with the needs of the times, carry out research and commercialization on NVIDIAs end-to-end AI platform, so as to create a sustainable agricultural future.
From sowing, transplanting, fertilization, spraying, harvesting and picking, almost all the work in the future can be completed automatically in the whole chain. Agricultural machinery automation, which was originally forced by the aging population, is now benefiting farmers.
Source: hyperai super neural editor: Mao Xinsi_ NBJS11624