Why It’s Time to Start Using Artificial Intelligence in Your Greenhouse

Luna by iUNU artificial intelligence

Luna by iUNU is a greenhouse AI platform using aerial robots on tracks that scan the entire production area several times a day using computer vision.
Photo courtesy of Sunrise Greenhouses

With today’s automated climate-control systems, managing the greenhouse environment and its systems has never been easier. The management of all equipment under one system, including heating, venting, and irrigation, provides optimal efficiency and precision in terms of systems management and data collection.

Advertisement

The development of these technologies allows growers to be more focused on their crops and provides control at their fingertips. Digital parameters are defined by crop level observations, which represent the modern-day dynamic between plants, the grower, and technology.

Increasing development of artificial intelligence (AI) has allowed the greenhouse sector to further improve efficiency and precision, thereby changing that dynamic. Thanks to the work of researchers and private companies, AI is allowing plants to communicate directly with climate-control systems, as well as the grower. It notifies them when there are issues within the crop relating to growth rate, pests, and disease. AI is not only helping growers identify when there is a problem, it is also predicting long-term effects of the issue via harvest forecasting.

Who Is Developing AI for the Greenhouse?

Top Articles
Highlights From Greenhouse Grower’s March 2024 Issue

Wageningen University and Research (WUR) is one of the academic institutions leading the way in the development of AI-controlled cultivation. In 2018, WUR hosted an autonomous growing competition where Microsoft took first place over several other teams, including a team of Dutch growers.

A host of private companies including Motorleaf, LetsGrow.com, and Illumitex already have commercially available platforms. Sunrise Greenhouses has been an early adopter of the technology, integrating AI in two areas of its production. The decision to start using AI was driven by the desire for improved efficiency and to help manage skilled labor shortages by expanding employee capacity.

The first system, Luna by iUNU, is a greenhouse AI platform using aerial robots on tracks that scan the entire production area several times a day using computer vision. Images are uploaded for analysis using algorithms customized to production requirements and translated for access on any device. Historical data is used to identify anomalies within the crop, alerting the grower or the climate-control system and providing an unprecedented level of oversight.

The second technology, Watchdog by Bold Robotic Solutions, is a production line AI system developed to monitor equipment for production issues and efficiency. Watchdog is currently installed on our potting line, monitoring pots moving from the potting machine to the tagger, transplanter, and placing robots. The system provides visual and audible alarms for issues such as an empty pot dispenser or pots that have fallen over. A series of sensors also allows the system to observe patterns and make timing corrections by controlling the equipment, thus removing repetitive corrective burdens from the operator.

What Are the Challenges?

In the upcoming years, there is no doubt the role AI will play in the evolution of the greenhouse industry. That being said, these game-changing technologies will not come without challenges, specifically during the early years of technology adoption.

While both solutions have different functions, they are based on the same self-learning technology that uses neural networks and machine learning. Like us, it takes time for the systems to learn patterns and crop cycles. As a seasonal grower with product cycles of up to two years in duration, patience is necessary. It takes time to collect the data for these systems to work and learn.

Greenhouse environments are also challenging for technology implementation due to broad temperature and humidity ranges, which influence both the electronic and mechanical components that contribute to their ongoing development. This can be a frustration for staff trying to complete their weekly plans.

AI solutions for greenhouse growers are still in their initial phases of development and have been mostly implemented by early adopters. As more players in the industry move toward the latest greenhouse technologies, we can expect the number of out-of-the-box AI solutions on the market to increase. For now, AI adopters will likely require some patience as the product learns the intricacies specific to their growing environment before it produces state-of-the-art results.

How Are Staff Trained?

The integration of these systems requires changes to processes, which can be disruptive to production, so flexibility and managing expectations is important among staff. When identifying new equipment, we involve staff in the process. When staff understand what we are trying to accomplish and can provide input, the transition is generally much smoother. People are more willing to adapt when they understand how the technology will make their lives easier. That said, providing people with training and, more importantly, ongoing support is key to the successful integration of these new technologies into your business.

AI digital crop surveillance on iPhone

The high level of granularity in digital crop surveillance can save money through early identification of problems related to pests, disease, moisture management, and climate. Photo courtesy of Sunrise Greenhouses

What Are the Benefits?

Having systems that continuously monitor equipment and cultivation allows staff to focus on less redundant tasks, which improves efficiency and the experience and quality of work. The high level of granularity in terms of crop surveillance has saved money due to the early identification of problems related to pests, disease, moisture management, and climate. This allows issues to be dealt with before they proliferate.

Working with iUNU, both our Production Manager and Sales Manager can access inventory in real-time, which has uncoupled departments in our facility that were previously dependent on one another for decision making. This is just one example of the increase in efficiency we are seeing as a result of these technologies.

Facilities that are focused on crop yields, such as vegetable growers, can benefit from AI developed by companies like Motorleaf, who offer yield-predicting products. They consider variables such as historical weather forecasts, nutrient ratios, daily temperatures, and humidity levels. These products train themselves to predict harvest yields with accuracies that are far superior to what is achieved by traditional forecasting methods.

4