Facebook AI Is Getting Into Agriculture

Facebook AI Is Getting Into Agriculture

With crop prices in the dumpster and the world’s population growing among a changing climate, artificial intelligence is becoming a life-saving measure for many farmers. From automated planting and harvesting to unmanned vehicles for cultivation and soil sampling, AI has begun to make it more cost efficient for producers to do their job.

One of the largest roadblocks is herbicides. According to a 2016 University of Illinois study, the chemical prices are on the rise and pose a big threat to a farmer’s bottom line.

“Growth rates of fertilizer, pesticide, and seed costs have been higher in years following 2006 than they were between 1990 and 2006. In 2015, the sum of fertilizer, pesticide, and seed costs were 48% of crop revenue, much higher than the 36% average from 1990 to 2006. These costs need to decrease, particularly if corn prices remain below $4.00. Otherwise, it will be difficult for revenues to be less than total costs.”

Recommended For You

Now, Facebook is getting into the farming game. Artificial intelligence co-maintained by the company, Pytorch, is being put to use by Blue River Technology — a subsidiary of John Deere — to improve weed control and bring down the costs of herbicide use.

“Farmers today face a huge challenge — feeding a growing global population with less available land,” said Alexandru Voica, Technology Communications Manager for Facebook in an email. “This, paired with variability inherent in farming, like changing weather conditions, and threats like weeds and pests, impact a farmer’s ability to produce food. We will share how PyTorch-enabled deep learning technology powers agricultural robots that make real-time decisions in the field; analyze images of crops in a matter of milliseconds; and support the building of ‘weed detection models’ for more efficient farming.”

The weed-controlled device works by integrating cameras, computer vision, machine learning and robotics to make an intelligent sprayer which drives through fields to target and spray weeds, leaving the crops intact, working essentially the same way facial recognition technology operates. The program was shown thousands of pictures of weeds until it was able to recognize the difference between the weeds and crops. According to Blue River Technology, it would save up to 90% of what is currently used for herbicides.

As it is pulled through the fields, the cameras analyze each frame and produce a pixel-accurate map of where the crops and weeds are. Once the plants are all identified, each weed and crop is mapped to field locations, while the robot sprays only the weeds.

The program PyTorch is an open source deep learning framework which enables developers and researchers to perform research and seamlessly deploy in production. Facebook AI researcher, Soumith Chintala, was a co-creator of the platform, and while it remains an open-source framework with a number of developers and researchers contributing to its growth, Facebook has said they will be dedicated resources to be a core maintainer of the platform.

It isn’t the first time that AI has been involved in the weed-control business. In 2018, Swiss company EcoRobotix released a solar-powered robot which could work for up to 12 hours detecting and destroying weeds. The company says the robot uses 20 times less herbicide than traditional methods. Also in 2018, the Danish company Agrinavia began developing a system called RoboWeedMaPS that would also automatically recognize and detects weeds in the field.

Published at Fri, 07 Aug 2020 15:55:23 +0000

Will You Soon Download Packaging Machine Controls from the Internet?

Rexroth’s new ctrlX automation platform leans into the Internet of Things age with software-smart controls and potential future “app store” offerings.

Led by the mantra that “mechanical engineering is based on software development,” Rexroth, a Bosch Co., has brought a new generation of automation software to market. The new system, ctrlX Automation, extends beyond the company’s hardware heritage in areas of hydraulic, electric, pneumatic, and mechanical drive and control technologies to offer a platform for complete automation solutions.

The new software-focused system brings traditional tools together with IT standards to bring control engineering into the age of smart manufacturing the Internet of Things (IoT), in keeping with a new corporate strategy we reported on last year. This relates to our May story, “3 Packaging Lines Improved by IoT Data,” but the company wanted to wait to disclose details pending its US market release…which is now official.

Dave Cameron, director of sales, presents latest update on Rexroth’s “Factory of the Future” vision.

Designed for machine builders, systems integrators, and end users with the resources to do their own control engineering, the system is initially targeted for “specific, focused applications,” namely robotics, material handling, and drives. In packaging, this makes it applicable for secondary and end-of-line realms. According to Dave Cameron, director of sales, future apps are in the works for additional packaging, assembly, printing, additive manufacturing, and machine tool apps.

Abbreviating applications to “apps,” incidentally, is one cue that the system is positioned for this era of IoT- and mobile-friendly deployments, as evident in the internet standards that largely define the system.

The software includes preconfigured apps that can be further customized. These include a PLC, motion controller, drive, human-machine interface (which can be web-based), monitoring dashboard, IoT gateway, and more. Apps in development include artificial intelligence (AI) and machine learning routines to monitor and detect variations in machines or processes. (See more apps here.)

Hardware designed for IoT-aware apps.

Rexroth says the system has “eradicated the traditional boundaries between machine control systems, IT, and the Internet of Things.”

Let’s start with a new generation of hardware control modules called ctrlX Core. Based on 64-bit multicore CPUs. The control modules are offered in three forms: a standalone ctrlX controller, on a control module integrated into a servo drive, or an industrial PC. These run Linux, with the IPC, which runs Windows and will also offer Linux. This offers scalability depending on the function, from real-time control to the greater overhead of workstation-level applications, as needs require. These support fieldbus options such as EtherCAT, Sercos, Ethernet IP, and Profinet. Support, in fact, spans common device protocols, interfaces, and communication standards, including support for more than 30 IT and IoT standards OPC UA (server and client) and MQTT, the leading machine-to-machine connectivity protocol well known for driving closer interfacing between operational and IT systems.

Software designed for IT inroads.

Cameron called the Core “the heart” of the new platform. That being the case, the system’s crown jewel software, ctrlX Works, is its digital heart of hearts. Claims that it reduces engineering time by 30% of 50% owes to the software’s large portfolio of preconfigured apps for common factory automation tasks. Users can customize them, develop their own (in seemingly any programming language), and more integrate third-party apps with relative standards-compliant ease.

Starting with programming, users can choose between their familiar, traditional tools as well as new methods. These include IEC 61131, PLCopen, or G-Code, or other conventional high-level or internet languages, which Rexroth says “liberate” teams from depending only on hard-to-find PLC specialists.

Software functions can be based on preconfigured or customized apps, combining traditional and newer languages: C++, Python scripting, or graphical languages, such as Blockly or Node Red, a low-code Javascript graphical environment. This variety helps users expand their skills (and recruiting) base and opens new possibilities for app development through GitHub, the leading open-source community for code-sharing.

The software for system design, maintenance and engineering can be accessed directly on the ctrlX CORE controller, by using any device with a simple web-browser.  This eliminates the need for separate engineering software on a different PC.

To show how web-based engineering works, Rexroth’s Brian Schmidt, application engineering supervisor, demonstrated the creation a 3f-axis motion control app using IoT-friendly Node Red, a low-code, web-based graphical programing environment for Javascript — which he also uses to build a web-based human-machine interface. “Most of the program flow is created graphically, but you can also write snippets of code,” he says, as he creates the program, jogging axes and the robot. He then tests the system using a 3D visualization app for feedback. Users can simulate their project before they download their actual automation programs to real-world hardware. He also created a pick-and-place application using different programming options. (See video below.)

Rexroth’s Brian Schmidt demonstrates web-based engineering by building a 3-axis motion control app (at the 5-minute mark), and shows its operation using a 3D visualization app. At the 10-minute mark, he shows another way to program a pick-and-place application, then gives a closer look at the control hardware running the demo.

App store, code sharing.

The apps are modular, and priced individually. “You just buy the software according to what you need for your application,” says Allen Tubbs, product manager Rexroth controls. Also in the works: a device portal for at-machine upgrades and an online app store. According to Tubbs, the system is designed to be “as open as possible. That means inviting customers and even other providers to develop solutions around our architecture.” He says the company currently has a developer portal, and plans to encourage GitHub participation, as well.

Assurances of future-proofing include extensive compliance with standards for data communications and legacy software migration/interfacing; secure encrypted communications based on IEC standards that will continue to evolve; and provisions for 5G and time-sensitive networking.

As end users gain experience with IoT data visibility, this type of system is poised to help companies extend their IoT implementations for greater machine, line, and plant performance. The data connectivity and visibility gained, Rexroth points out, will lead to more effective forays in machine learning and AI, up to and including lights-out manufacturing.

Published at Fri, 07 Aug 2020 15:33:45 +0000