Remote Control of Greenhouse Vegetable Production with Artificial Intelligence—Greenhouse Climate, Irrigation, and Crop Production

Abstract

The global population is increasing rapidly, together with the demand for healthy fresh food. The greenhouse industry can play an important role, but encounters difficulties finding skilled staff to manage crop production. Artificial intelligence (AI) has reached breakthroughs in several areas, however, not yet in horticulture. An international competition on “autonomous greenhouses” aimed to combine horticultural expertise with AI to make breakthroughs in fresh food production with fewer resources. Five international teams, consisting of scientists, professionals, and students with different backgrounds in horticulture and AI, participated in a greenhouse growing experiment. Each team had a 96 m2 modern greenhouse compartment to grow a cucumber crop remotely during a 4-month-period. Each compartment was equipped with standard actuators (heating, ventilation, screening, lighting, fogging, CO2 supply, water and nutrient supply). Control setpoints were remotely determined by teams using their own AI algorithms. Actuators were operated by a process computer. Different sensors continuously collected measurements. Setpoints and measurements were exchanged via a digital interface. Achievements in AI-controlled compartments were compared with a manually operated reference. Detailed results on cucumber yield, resource use, and net profit obtained by teams are explained in this paper. We can conclude that in general AI performed well in controlling a greenhouse. One team outperformed the manually-grown reference.

Cherry Tomato Production in Intelligent Greenhouses—Sensors and AI for Control of Climate, Irrigation, Crop Yield, and Quality

“…Recently a benchmark experiment has been conducted to use artificial intelligent (AI) algorithms to optimize net profit of a cucumber crop in a greenhouse experiment during the first Autonomous Greenhouse Challenge in 2018. In that experiment the winning AI algorithm outperformed the human decisions of experienced growers [ 58 ]. Crop production (class A: commercially sellable fruits) was increased by 6% and net profit by 17% compared to the growers who acted as a reference.…”

“…The challenge was designed to make further breakthroughs in fresh food production with fewer resources using AI algorithms for and automatic and remote control of a greenhouse crop production. While the first experiment [ 58 ] was simpler with only a 3–4 months control of cucumber production, this second experiment was more complex to prove the value of AI control over a longer six months period. A different crop had to be grown, cherry tomatoes require more complex control since it can be controlled not only on yield but also on product quality.…”

“…Other challenges were added such as no fixed product prices, but prices were dependent on fruit quality and fruit quality was dependent on nutrient control, nutrient control was an offtrade of yield (more income but lower prices) and product quality (high prices but lower yield). Comparable to the first experiment [ 58 ], five multi-disciplinary international teams automatically controlled their own greenhouse compartment remotely, a sixth team of local experienced human reference growers acted as manually controlled reference. The goal of the experiment was to maximize net profit by realizing high yield and product prices and minimize resource use and costs.…”

Environment Monitoring of Rose Crops Greenhouse Based on Autonomous Vehicles with a WSN and Data Analysis

“…In this work, data balancing ( Section 4.2 ) is implemented to avoid over-training regarding majority-class samples. In addition, a prototype selection technique ( Section 4.3 ) is used in order to eliminate data that does not provide important information to the classifier [ 26 ]. Finally, a comparison of the classical supervised classification criteria ( Section 4.4 ) is made to choose the appropriate algorithm that maintains a high compromise between the classification performance and the computational cost that the decision represents.…”

IoT-Based Sensor Data Fusion for Determining Optimality Degrees of Microclimate Parameters in Commercial Greenhouse Production of Tomato

“…This is particularly important in commercial scale cultivation under closed-field environments, where disease can be spread in a shorter time and production failures impose serious setbacks. In this regard, wireless sensors and remote monitoring-and-control instrumentation that benefits from the concept of the Internet-of-Things (IoT) have been deployed in smart farming to help growers stay competitive at the market [ 5 , 6 ]. Wireless sensing becomes more demanding in greenhouse applications due to their high flexibility and functionality for real-time monitoring.…”