IoT in agriculture: farming gets ‘smart’

Solving problems such as the impending global food shortage will require concerted effort from various countries, organizations and groups from across the world. Safely incorporating some of these innovative techniques to improve yield, decrease loss and increase process efficiencies will go a long way in alleviating this immense challenge.

strawberries
Sharon Mollerus (CC BY 2.0)

We continue our theme of looking at different industry verticals, big and small, that are getting the IoT treatment. Previously we have looked at the largest IoT deployment, the Smart Electric Grid, as well as some innovative IoT implementations that are transforming healthcare. Today, it is time to look at Agriculture. While it might not seem to be a ‘cool enough’ vertical to get the ‘Smart Treatment’, this couldn’t be further from the truth.

Why is it vital to even implement Smart Farming? Two reasons – carbon dioxide and human population growth. The increase in CO2 in the atmosphere leads to lower production, while the steady growth in human population leads to increased demand. Many predict that unless we increase our food production, we are very close to a global food shortage. Since it is not easy to dramatically increase arable land size, one way to achieve our goal is to improve the yield per unit area (or lower waste). Here is where IoT comes in.

IoT initiatives in agriculture

Despite images people may have of farmers toiling away, the reality is farming has become a very advanced mechanized industry. Large farming implements – seeders and harvesters, have led to a high level of semi-automated farming processes. It is true that these are currently piloted by human drivers, but autonomous vehicles existed in fields long before today’s pseudo self-driving car. Large corporations have replaced small farms, at least in terms of ownership. There are already several ‘connected’ initiatives in use today – from soil monitoring to irrigation sensors.

One example of this is tracking the health of cattle. Cattle farms and ranches usually stretch over a large land area, making it difficult to monitor the whereabouts of grazing animals without human involvement. Using tracking collars, one can find the location of these animals in real time. Then, a data storage system can record this information in a database to ultimately form a baseline model of their movements during a given time period. Applying intelligent algorithms on these patterns helps us identify if the cattle’s movements are irregular, of if one or more animals are separated from the herd. This usually occurs if they are sick or injured. This solution can easily be implemented by small IoT trackers that communicate over an IoT network like Wi-SUN or other WANs. One could then have networking towers distributed across the fields to cover a large area. This information is then exposed to the farmer or rancher via a web portal or smartphone application, thus making it easy from them to consume it.

Another area of IoT use in farming is the utilization of drones to improve crop health. Disease, and the ease of which disease spreads amongst crops, is a real cause for concern as this directly impacts crop yield. Preventing disease is the ideal scenario, but an easier and more realistic goal is to immediately identify, isolate and remove affected crops. A novel way to do this is by using surveillance drones. These drone groups are housed at a base station on a farm, from which they make automated, periodic patrols to capture imaging data on the crop. Using computer vision/image recognition algorithms, one can determine which areas on a farm that are adversely affected. Tagged images are correlated with drone GPS provide pinpointed information to the farmer. This information can be assimilated from various drone recordings, analyzed and then reported to the farmer, who can then take remedial action. The entire process from launching the drones, navigating them, capturing data, analyzing it and reporting on it is all automated.

Precision farming is another area where the use of connected sensors is rampant – we are even seeing these devices enter the consumer space. Remote, battery operated soil sensors collect information regarding its nitrogen content and report this back on a periodic basis; irrigation sensors measure the water level and automatically inform the irrigation and sprinkler system; flood sensors monitor water levels as well and can be set to automatically turn off the valves that are responsible for over-watering the plants and simultaneously send a notification email to a specified email; finally, a freeze sensor can sense, and automatically alert users, when weather conditions lead to frost that might harm sensitive plants. For homeowners, there are now smart sprinkler controllers that use the home’s Wi-Fi network to expose control functionality via a smartphone app. One can set schedules, remotely toggle the sprinklers and even control the water flow.

Why do we need security?

Thus far the use cases we’ve mentioned seem relatively simple and their threat surfaces benign – but this can lead to a false sense of security. One of the goals of smart agriculture is to automate manual processes and ensure low touch practices. However, this can make them easy targets to hackers.

Often, these systems are running on unmonitored networks, due to which any attempted or successful breaches go unreported. Worse, since this sector is traditionally not cybersecurity aware – like banking – security design is not incorporated into the solution requirements. Hackers can easily gain access to irrigation control systems of a plant and then manipulate them maliciously, or demand ransom to relinquish control. Administration of pesticides that are carefully controlled can be manipulated without the farmer knowing. Finally, these internet-connected systems can be used to gain access to other connected third-party systems and ultimately become part of a botnet. The avenues and motives for attack are plenty!

So how do we solve these security problems? The onus lies on the smart device makers for some self-regulation and adherence to security best practices if not to security standards. As we’ve said before, there is no need to reinvent the wheel. When it comes to security, the principles that secure the internet today, will secure the internet of things tomorrow. The advice is simple – stand on the shoulders of giants, use tried and tested security methodologies, talk to security experts and make security by design a core tenet of your solution. Public Key Infrastructure is like a Swiss Army knife – it will help identify your devices, protect communications from eavesdropping and safeguard your secrets. Combine this with secure elements to encrypt and securely store your data and keys and you now have a solid security foundation built upon strong identity, authentication and trust.

Solving problems such as the impending global food shortage will require concerted effort from various countries, organizations and groups from across the world. Safely incorporating some of these innovative techniques to improve yield, decrease loss and increase process efficiencies will go a long way in alleviating this immense challenge.

This article is published as part of the IDG Contributor Network. Want to Join?

Join the Network World communities on Facebook and LinkedIn to comment on topics that are top of mind.
Now read: Getting grounded in IoT