According to the U.S. Department of Agriculture's Bureau of Agricultural Research, insect pests like The Scarab beetle in Japan cost agriculture more than $100 billion a year. Coupled with the plant diseases that these insects can transmit, arthropods cause 40% of the world's agricultural production losses each year. FarmSense is an agri-tech startup based in Riverside, California, that is trying to solve the problem of insect pests. The company has created optical sensors and a new classification system based on machine learning algorithms to identify and track insects in real time. The key here is: real-time information.
They claim that the real-time information provided by their sensors allows for early detection, allowing for the timely deployment of pest management tools such as pesticides or biological control. The mechanical trappers currently used for monitoring may only generate significant intelligence after the bugs arrive 10 to 14 days later.
Eamonn Keogh, co-founder of FarmSense, said: "The adults of these bugs only live for 5 days, so when you know you have a problem, the problem has already arisen and now it's a bigger problem. If you know the problem in real time, you can localize the intervention, do it in only one place, and the results will be much better, saving pesticides, saving labor, and leaving crops unharmed. ”
How they can provide information that is critical to achieving these better outcomes is a bit complicated.
The company is currently testing and researching in almond orchards in Southern California, thanks to the Small Business Innovation Research Fund. Their latest sensor, called FlightSensor, is best understood when considering where Keogh got the idea from: James Bond and Cold War-era espionage.
Keogh explained how Russian spies used lasers, positioned on glass window panes, to capture vibrations caused by people's voices. A sensor will then translate the information, providing rough intelligence about what's going on in the room. "With the same thought, I imagined what would happen if a bug flew over a laser... You will hear only the sound of bugs and nothing else. ”
However, instead of reading the vibrations, FlightSensor uses light curtains and shadows within a small channel where insects are attracted in. On one side of the sensor is a light source and on the other side is an optical sensor. Sensors measure how much light is blocked, or how much light can pass through, as insects fly inside. This data is converted into audio and analyzed by machine learning algorithms in the cloud.
According to FarmSense, the sensor is designed to look like an old-fashioned analog device and does not receive ambient noise, such as wind or rain, according to FarmSense.
Keogh said: "The quality of the signal is very clear, it is deaf to the ambient sounds that are usually heard in the field. It's basically a different mode of hearing, but when you put on headphones to listen to audio clips from the sensor, it sounds like a mosquito or a bee flying around. ”
Keogh, a professor in the Department of Computer Science and Engineering at the University of California, Riverside, specializes in data mining and works on FarmSense's novel machine learning algorithms for identification purposes. Assisted in the development and deployment were entomologists and field experts, including co-founder Leslie Hickle.
The company's CEO, Shailendra Singh, has developed systems for wireless and cellular networks as well as security – working on the hardware side. He offers a working price point for each sensor, which will be billed seasonally for $300.
The impact of this technology is clear. For farmers, real-time information about insects is not only important for their financial security, but also makes it possible for them to preserve and protect critical resources, such as soil health.
But FarmSense claims it wants to empower rural farmers because they say the damage caused by insects affects them particularly greatly.
However, the price of $300 per sensor per quarter is high, which poses a potential risk to the adoption of the technology and therefore to the technology's ability to solve the pest problem in the first place.
Michael Carter, director of the USDA-funded Feed the Future Innovation Lab on Markets, Risk and Resilience and distinguished professor of agricultural and resource economics at the University of California, Davis, said one of the most difficult things for small-scale farmers is managing risk.
"Risk can keep people poor. It hinders investment in technologies that can raise average revenues because the future is unknown," Carter said. "People with less wealth obviously don't have a lot of savings, but they can't risk savings to invest in things that might raise their incomes, which can also lead to starvation of their families."
However, he is optimistic that technologies like FlightSensor could ease small-scale farmers' fears of investment, especially if the technology is paired with insurance to further protect them.
This technology also raises this question. Is real-time identification really the best option for pest management? Andrew Lieb, a research entomologist at the U.S. Department of Agriculture's Forest Service, said this may not be. He explained that the main drivers of invasive insects — often the most destructive to agriculture and forests — are travel and trade.
He expressed optimism about technology as a way to control insects, but ultimately decided that the best strategy was to attack the problem earlier. We should address the current import and export laws and how to deal with products to remove pests.
Despite these concerns, there is no doubt that FarmSense's technology is ready to make an impact. Even beyond addressing farmers' financial insecurity and threats to our global food chain, it could prove useful in tracking and disseminating critical information about disease-borne insects, such as mosquitoes.
Given that non-native insect invasions are expected to increase by 36 percent by 2050, and that growing population sizes will put more pressure on food production, innovative technologies like FlightSensor are very welcome to advance people's ability to understand and respond to threats thoughtfully.
As Carter said of all the possible ways that agricultural technology can still benefit agriculture, "we need to be creative on these fringes."