An Update To Solving The Impending Food Crisis

As I reported earlier, the the Global Harvest Initiative (GHI) announced that the demand for food will outpace food production by 2050 when the world’s population is estimated to reach 9 billion people. The report according to Margaret Zeigler, executive director of GHI, is “a call to action.”

“Countries need to prioritize agriculture and the growing of food in more sustainable methods,” Zeigler argued. “If we don’t start now, we’ll have a problem sooner, even by 2030.”

In response, members of the GHI, including such multinational companies as Dupont, Monsanto, John Deere, are investing in mechatronic solutions. Profiled below are three ground-breaking technologies which are changing the agrarian landscape:



To counter rising demand, compounded by the growth of urbanization, farms are moving from large, inefficient horizontal rural plains to indoor city-based agricultural factories. Instead of relying on natural sunlight, these walled gardens utilize new LED lighting technology that beams down on rows of crops stacked one on top of another. In addition to increasing production and reducing waste, indoor vertical farming also eliminates runoff from pesticides and herbicides.

In Japan, the farming company Spread is taking vertical farming to a new level by launching the world’s first robotic farm that is planning to harvest 30,000 heads of lettuce a day. Spread’s new automation technology will not only produce more lettuce, it will also reduce labor costs by 50%, cut energy use by 30%, and recycle 98% of water needed to grow the crops.

“The use of machines and technology has been improving agriculture in this way throughout human history,” J.J. Price, a spokesperson at Spread. “With the introduction of plant factories and their controlled environment, we are now able to provide the ideal environment for the crops.”


Spread plans to expand its plant factories across Japan. Here in the United States, fears of farm labor shortages as a result of the President-Elect’s plan to close the Mexican border has accelerated interest in robotic farmhands. Already, there has been a declining number of unauthorized immigrants in the U.S. since its peak in 2007. This is due to increased job opportunities in Mexico, as well as tighter U.S. border patrols implemented by the Obama administration.

To counter this trend SRI Ventures, a startup incubator of the Menlo Park research and development firm SRI International, announced in August their investment in Abundant Robotics Inc. to automate the harvesting of apples. The USDA estimated that 2016’s apple crop exceeded 245 million bushels – making it one of the most popular grown, and exported, foods in America. Apples, due to their thick density, are also one of the easiest foods for robots to harvest vs. soft-skinned tomatoes. Yet, apples are still being harvested today in the same centuries old hand-picked manner. Abundant Robotic aims to change that with its vacuum-picking robot (above).

According to Abundant’s CEO, Dan Steere, their vacuum-end effectors are able to pick at a rate of one apple per second without damaging the fruit. This is currently faster than the industry worker average of one apple every 2 seconds. To do this Steere’s invention utilizes computer vision software to recognize ripe apples before sucking them off the tree. The robot is pulled by a tractor that provides both storage for the harvest, and power for the robot.

Steere explained, “seeing fruit and picking it without damaging it is the big engineering challenge. If you bruise or cut the fruit it loses its value.”

Manish Kothari, president of SRI, confirmed the complexities of the system, “It’s a very non-trivial engineering challenge. To detect apples very precisely you have to see down at the millimeter level in real time. That requires software, and on the hardware side, chips that allow you to do real time image processing on the fly.”

The team spun out of SRI and partnered last year with the Washington Tree Fruit commission to bring automation to the field. According to the press release, most of the intellectual property was developed at SRI in partnership with Carnegie Mellon University. Abundant Robotics aims to have its autonomous apple pickers on orchards within the next two years.

Abundant’s deployment plan fits into a larger timeline of the deployment of fresh fruit picking robots. According to research firm IDTechEx, “the progress of robotic deployments have been hampered by the stringent technical requirements of farming. The vision system needs to detect fruits inside a complex canopy whilst the robotic arms needs to rapidly, economically and gently pick the fruit.” The report cites a limited number of commercial implementations, including one from Agrobot below that required strawberries to grown in an unusual hanging fashion.


According to the IDTechEX report, wide-adoption of robotic harvesters will start around 2020 (see chart below). The researchers estimate that in the next five years with the development of quality robotic arms and neural network systems, roboticists will be able to offer systems with greater efficiency and cost savings than presently available with human labor.

Screen Shot 2016-12-15 at 7.26.26 PM.png


Last July, PwC projected that the market for remote-sensing services by drones and satellites within the agriculture industry will exceed $30 billion. Drone technology offers the promise of enabling farmers better planning and management by utilizing the availability of real-time data. PwC’s report listed six ways aerial drones are being utilized throughout the crop cycle; below is an annotated summary:

Screen Shot 2016-12-17 at 9.47.53 PM.png

Soil and field analysis: Drones can be instrumental at the start of the crop cycle. They produce precise 3-D maps for early soil analysis, which is useful in planning seed planting patterns. After planting, drone-driven soil analysis provides data for irrigation and nitrogen-level management.

I happen to be the advisor of a leading drone agronomic company, GreenSight Agronomics, that is providing a turnkey analytics service for the entire turfgrass ecosystems from farmers and service providers. According to James Peverill, CEO, “GreenSight’s proprietary drone, sensors, and plant-health analytics provide turfgrass managers with actionable information on the health of their turf so they can make better decisions on when to apply fertilizer, pesticide, and optimize their irrigation systems.”

Planting: Startups have created drone-planting systems that achieve an uptake rate of 75 percent and decrease planting costs by 85 percent. These systems shoot pods with seeds and plant nutrients into the soil, providing the plant all the nutrients necessary to sustain life.

Crop spraying: Distance-measuring equipment—ultrasonic echoing and lasers such as those used in the light-detection and ranging, or LiDAR, method—enables a drone to adjust altitude as the topography and geography vary, and thus avoid collisions. Consequently, drones can scan the ground and spray the correct amount of liquid, modulating distance from the ground and spraying in real time for even coverage. The result: increased efficiency with a reduction in the amount of chemicals penetrating into groundwater. In fact, experts estimate that aerial spraying can be completed up to five times faster with drones than with traditional machinery.

Companies like GreenSight are working to provide pinpoint geolocation data to address crop issues below ground before they spread to the plants above. By working working with vendors like agrochemical and irrigation suppliers, GreenSight is reporting back to them the most effective treatments and water conservation methods.

Irrigation: Drones with hyperspectral, multispectral, or thermal sensors like GreenSight can identify which parts of a field are dry or need improvements. Additionally, once the crop is growing, drones allow the calculation of the vegetation index, which describes the relative density and health of the crop, and show the heat signature, which is the amount of energy or heat the crop emits.

Crop monitoring: Vast fields and low efficiency in crop monitoring combine to create farming’s largest obstacle. Monitoring challenges are exacerbated by increasingly unpredictable weather conditions, which drive risk and field maintenance costs.


These techniques are being implemented on a wider-scale by satellite companies like PlanetWatchers that monitor forestry assets worldwide. PlanetWatchers’ proprietary algorithms are able to see through clouds to report on crop health, illegal logging, and audit conversation. PlanetWatchers will also partner with drone operators to provide precise data to carry out missions within these vast areas that would be impossible for quadcopters to survey.

Health assessment: It is essential to assess crop health and spot bacterial or fungal infections on trees. By scanning a crop using both visible and near-infrared light, drone-carried devices can identify which plants reflect different amounts of green light and NIR light. This information can produce multispectral images that track changes in plants and indicate their health. As an example, GreenSight alerts field managers with actionable data to remedy planet diseases before they are discoverable to the naked eye.

One of the most novel approaches of drones to determine the health of crops was demonstrated at Ted a couple of years ago by Professor Vijay Kumar of the University of Pennsylvania (above). Utilizing proprietary swarm mechanics, Kumar’s lab illustrates how one could use multiple drones to estimate crop yields prior to harvest. The drones fly in unison and actively report the data to the cloud.

berry robots.gif

The promise of solving the food crisis will be driven in large measure by the public acceptance of robots handling its food supply. While most predict the impact to be positive, I would be remiss to overlook the loss of human jobs. However, I leave that discussion for our next #RobotLabNYC Community MeetUp on March 2nd!

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: