Revolutionizing Agriculture with Computer Integrated Systems

Imagine a world where farming is as precise and efficient as manufacturing cars. This is the promise of computer-integrated systems in agriculture and horticulture. These systems, often referred to as “Intelligent Plant Factories,” leverage advanced technology to optimize the cultivating process, ensuring maximum yield and efficiency. Let’s dive into how these systems work and their incredible potential.

The Rise of Computer-Integrated Systems in Agriculture

1. Environmental Control

Managing the environment within a greenhouse is a complex task due to fluctuating sunlight and other variables. Traditional methods focused on maintaining constant conditions, but newer approaches emphasize optimal control, adapting to changes for the best results.

• Key Techniques:
  • Micro-computers and PID Algorithms: These technologies help maintain the perfect balance of light, temperature, humidity, and CO2.
  • Nutrient Management: Advanced sensors and control algorithms ensure plants receive the right nutrients at the right time.

2. Adaptive, Optimal, and Fuzzy Control

For more flexible and cost-effective environmental control, adaptive systems are key. These systems adjust settings based on real-time data, ensuring plants thrive under varying conditions.

• Adaptive Control: Adjusts to sudden changes in the environment.
• Optimal Control: Minimizes costs while maximizing plant health.
• Fuzzy Control: Uses artificial intelligence to handle complex nutrient management.

3. Mechanization and Automation

The labor-intensive nature of traditional farming is being transformed by automation. Robots and automated systems are increasingly handling tasks like seeding, transplanting, and harvesting, which boosts efficiency and reduces labor costs.

• Greenhouse Automation: Robots take over repetitive tasks.
• Micropropagation Robots: Used in tissue culture, these robots enhance precision and efficiency.

Intelligent Plant Factories: The Future of Farming

Plant Factory Systems

There are two main types of plant factories:

1. Fully Controlled Systems with Artificial Light: These systems create an ideal environment for plant growth, much like a large-scale growth chamber.
2. Solar-Based Systems: Similar to greenhouses but with enhanced control over environmental conditions.

These factories can accelerate plant growth significantly—up to five times faster than traditional methods—by providing continuous light, optimized CO2 levels, and perfect temperature and nutrient conditions.

Process Optimization with Plant Responses

To truly optimize plant growth, it’s essential to understand how plants respond to environmental stresses. This involves measuring various plant responses, from growth rates to photosynthesis efficiency.

• Speaking Plant Approach: This concept focuses on understanding and responding to the needs of plants based on scientific data and advanced algorithms.

Actionable Tips for Implementing Computer-Integrated Systems

1. Invest in Quality Sensors: Accurate data collection is crucial for maintaining optimal growing conditions.
2. Utilize Adaptive Control Systems: These systems will help manage sudden changes in the environment.
3. Incorporate Automation: Reduce labor costs and increase efficiency by using robots for repetitive tasks.
4. Monitor Plant Responses: Regularly check plant health indicators to adjust environmental conditions promptly.

Summary for Social Media and Infographics

Key Points for Instagram Reels and Canva Infographics:

• Introduction: Highlight the importance of technology in modern farming.
• Environmental Control: Explain how sensors and algorithms optimize conditions.
• Adaptive Systems: Showcase the benefits of flexibility and cost reduction.
• Automation: Emphasize the role of robots in reducing labor.
• Plant Factories: Describe the two types and their benefits.
• Actionable Tips: Provide quick, practical steps for farmers to implement these systems.

With these insights, agriculture enthusiasts and farmers can better understand and harness the power of computer-integrated systems to revolutionize their practices and achieve remarkable efficiency and productivity.

B. Approach from Information and Knowledge Processing

Almost all kinds of data can now be processed in real-time with the aid of computer systems. Information processing, including analog electric signals, can easily be done using software for digital signal processing. One notable application of image processing owes much to the Fourier transform, one of the most popular algorithms in digital signal processing. Techniques in image processing have elucidated plant responses in two dimensions. Additionally, NMR-CT, traditionally used in medical diagnosis, has begun to be utilized in physiological plant ecology , a development that would not be possible without the fast Fourier transform. Sensors with processing capabilities are also expected to enhance plant response detection significantly.

The advent of computers capable of processing human knowledge and making decisions has led to practical applications of artificial intelligence (AI) and knowledge processing. Systems that can make decisions based on input data, often referred to as AI, include fuzzy logic and neural networks, both found to be effective in various contexts. These advanced processing techniques should be integrated into the speaking plant approach.

C. Approach from System Identification and Control Engineering

Despite the extensive publication on the identification, modeling, and control of non-biological systems, there is a dearth of literature focused on plant growth. Plant or crop growth is closely linked to photosynthesis, which is affected by numerous environmental stresses, with CO2 concentration and light intensity being the most critical factors. Consequently, CO2 uptake in response to environmental stresses is a primary target for identifying and modeling environmental control systems. This has been partly identified using spectral analysis and the method of least squares.

1. Spectral Analysis

Spectral analysis has been applied to the photosynthesis of sunflowers affected by light intensity . The method involves describing light intensity as a signal x(t)x(t)x(t) and net CO2 uptake as y(t)y(t)y(t) in the time domain. The Fourier transform of light intensity is X(f)X(f)X(f), and the Fourier transform of net CO2 uptake is Y(f)Y(f)Y(f) in the frequency domain. Assuming a linear filtering relation between input and output based on g(t)g(t)g(t) or G(f)G(f)G(f) (Figure 1), spectral analysis is particularly effective in systems where the input and output are random variables. Power spectrum calculations using the Fast Fourier Transform (FFT) algorithm are shown in Figure 2.

2. Method of Least Squares Based on ARMA

The least squares method, based on ARMA (Auto Regressive Moving Average), involves input signal u(t)u(t)u(t), output signal y(t)y(t)y(t), and noise v(t)v(t)v(t), assuming v(t)v(t)v(t) and u(t)u(t)u(t) are statistically independent; The input-output relation is derived as follows:

y(t)=B(z−1)⋅u(t)A(z−1)+v(t)y(t) = \frac{B(z^{-1}) \cdot u(t)}{A(z^{-1})} + v(t)y(t)=A(z−1)B(z−1)⋅u(t)​+v(t)

where A(z−1)A(z^{-1})A(z−1) and B(z−1)B(z^{-1})B(z−1) are polynomials in the backward shift operator z−1z^{-1}z−1:

A(z−1)=1+a1z−1+…+anz−nA(z^{-1}) = 1 + a_1 z^{-1} + \ldots + a_n z^{-n}A(z−1)=1+a1​z−1+…+an​z−n B(z−1)=b0+b1z−1+…+bnz−nB(z^{-1}) = b_0 + b_1 z^{-1} + \ldots + b_n z^{-n}B(z−1)=b0​+b1​z−1+…+bn​z−n

Thus, the system is described and identified using vector and matrix representations of input and output data (Figure 3).

3. Group Method of Data Handling (GMDH)

The dynamic model of CO2 uptake can also be identified using GMDH, a method effective for identifying nonlinear systems of unknown structure . Developed by Ivakhnenko, GMDH involves combining input variables with quadratic polynomials and selecting optimal representations through successive layers (Figure 4).

4. Applications of Artificial Neural Networks to Identification

Applications of neural networks to system identification have been explored, where the input-output relationship is modeled based on ARMA The neural network structure is built from historical input and output data, with layers including input, hidden, and output layers (Figure 5). The network’s learning procedure, based on error back-propagation, adjusts weights and biases to minimize the error between observed and calculated values.

V. Computer Support System for Control, Horticultural Operation, and Management

The application of computers has broadened significantly, with their introduction into decision support systems for horticultural operation and management, as well as real-time control systems. AI, especially expert and diagnostic systems, has become increasingly integral.

A tomato cultivation support system, combining environmental control and artificial intelligence via computer networks, was found effective for large-scale plant growth factories, Expert systems for disease and pest diagnosis, as well as control system adjustments, have also been developed

Future advancements may involve expert systems for managing plant growth factories based on strategic decisions, aligning with the concept of Computer Integrated Manufacturing (CIM) in agriculture

CONCLUSION

The integration of computer systems into plant growth factories and micropropagation processes offers significant value enhancement. Analogous to the advancements in the chemical process industry, including process automation (PA), factory automation (FA), and computer integrated manufacture (CIM), the introduction of similar systems in agriculture and horticulture is expected to substantially boost productivity. Greenhouses and plant growth factories, which closely resemble chemical process industries, stand to gain particularly from these technological improvements.

A comprehensive computer integrated system, consisting of various specialized computers for tasks such as communication and networking (LAN), environmental control, nutrient control, factory automation, system identification and optimization, horticultural operation and management support, and expert and diagnostic knowledge processing, is essential for the industrialization of agricultural production. This multi-faceted system, as proposed in this paper, represents a logical and crucial step toward the realization of an “Intelligent Plant Factory.”

parameters of the greenhouse environment to meet the crop’s needs optimally. The challenge lies in balancing multiple and sometimes conflicting goals: promoting healthy plant growth, preventing diseases, and minimizing energy consumption. A detailed understanding and integration of both external and internal factors are crucial in this process.

B. Time Decomposition

To effectively manage the greenhouse climate, it is essential to break down the time into manageable units. This involves considering both the daily cycle and the entire growing season.

1. Daily Cycle:
   • Morning (Sunrise to Midday): Focus on gradually increasing temperature and ensuring adequate ventilation.
   • Afternoon (Midday to Sunset): Maintain optimal temperature and humidity levels while monitoring plant responses to the peak sunlight and heat.
   • Night (Sunset to Sunrise): Maintain minimum temperature setpoints to conserve energy while preventing frost or cold stress.
2. Growing Season:
   • Initial Growth Stage (November to December): Emphasize soil and air heating to support seedling establishment and early growth.
   • Mid-Growth Stage (January to March): Balance between maintaining optimal temperatures and managing energy costs, as this is a critical growth period.
   • Late Growth and Harvest Stage (April to June): Focus on maintaining consistent temperatures and adequate ventilation to ensure fruit quality and prevent diseases.

C. Decision Process Order

The decision-making process for climatic setpoints follows a logical sequence:

1. Assess External Conditions: Evaluate outside air temperature, solar radiation, and wind speed.
2. Monitor Internal Conditions: Check air temperature, soil temperature, saturation deficit, and plant physiological indicators.
3. Setpoint Adjustment: Determine the appropriate setpoints for air heating, soil heating, and ventilation based on current and forecasted conditions.
4. Evaluation and Optimization: Continuously monitor plant responses and environmental conditions, making necessary adjustments to optimize growth while conserving energy.

D. Translating Knowledge into Constraints and Rules

The empirical knowledge gathered from experienced growers can be translated into a set of constraints and rules that guide the decision-making process:

1. Temperature Management:
   • Air Heating Setpoint: Maintain air temperature between X°C and Y°C depending on the time of day and growth stage.
   • Soil Heating Setpoint: Ensure soil temperature remains above Z°C to support root development.
2. Ventilation Control:
   • Aeration Setpoint: Open roof windows if air temperature exceeds W°C or if humidity levels rise above acceptable limits.
   • Saturation Deficit: Adjust ventilation to maintain a balance between temperature and humidity, preventing excessive moisture that can lead to disease.
3. Plant Health Indicators:
   • Disease Symptoms: Modify temperature and humidity setpoints if signs of disease are detected to create less favorable conditions for pathogen development.
   • Vigor and Growth: Adjust setpoints to either stimulate or temper growth based on the observed vigor of the plants.

IV. ARTIFICIAL INTELLIGENCE APPROACH IN SERRISTE

The SERRISTE system employs artificial intelligence techniques to handle the complexity of the setpoint determination problem. By using an object-oriented programming approach, SERRISTE encapsulates domain-specific knowledge and decision-making processes within a coherent framework.

1. Knowledge Representation: The system explicitly represents empirical knowledge as constraints and rules, enabling a structured approach to decision-making.
2. Constraint Satisfaction Problem (CSP): SERRISTE solves the CSP by assigning values to variables (setpoints) within the constraints derived from empirical knowledge.
3. Preference Criteria: Once acceptable solutions are identified, they are evaluated and ranked based on economic and growth performance criteria.

V. PRELIMINARY EVALUATIONS AND FUTURE DEVELOPMENTS

Preliminary evaluations of SERRISTE indicate its potential to enhance greenhouse management by providing more accurate and dynamic setpoint adjustments. Future developments may include integrating additional environmental factors such as CO2 concentration and nutrient management, expanding the knowledge base with more empirical data, and refining the AI algorithms to improve decision accuracy and efficiency.

By combining empirical knowledge with advanced artificial intelligence techniques, systems like SERRISTE can significantly improve the productivity and sustainability of greenhouse farming operations, ensuring better resource utilization and higher crop yields.

on the variables of the problem and then applies a search algorithm that operates on the resulting domains. The search procedure works by systematically trying possible values for each variable, backtracking when a constraint violation is detected.

In SERRISTE, the search is organized according to the structure of the variable clusters defined earlier. Each variable cluster corresponds to a subtree in the evaluation tree, and the search proceeds by instantiating the variables in the current cluster before moving to its descendants.

The search algorithm can be summarized as follows:

1. Initialization: Start with the root cluster of the evaluation tree. Apply filtering to reduce the domains of the variables in this cluster.
2. Variable Selection: Select a variable from the current cluster to instantiate. This selection is based on predefined criteria, such as the difficulty of instantiation or user preferences.
3. Value Assignment: Assign a value to the selected variable from its domain. The choice of value can be guided by heuristics to improve efficiency.
4. Constraint Checking: Check if the current partial assignment satisfies all constraints involving the instantiated variables. If any constraint is violated, backtrack to the previous step and try a different value for the variable.
5. Cluster Advancement: If all variables in the current cluster are instantiated and consistent, move to the next cluster (a child in the evaluation tree) and repeat the process from step 2.
6. Backtracking: If no valid value can be assigned to a variable in the current cluster, backtrack to the parent cluster and try a different value for the previously instantiated variable.
7. Termination: The process continues until either a complete and consistent assignment for all variables is found (a solution) or all possible assignments have been exhausted (no solution).

The above process ensures that all constraints are respected and allows for the systematic exploration of possible solutions. Additionally, the use of variable clustering and domain filtering significantly reduces the computational effort required, making the search process more efficient and easier to understand.

D. Global Functioning of SERRISTE

The global functioning of the SERRISTE system can be divided into several phases:

1. Preprocessing Phase:
   • Knowledge Representation: Identify variables, domains, and constraints.
   • Clustering: Organize variables into clusters based on constraint-induced dependencies.
   • Tree Construction: Build the evaluation tree from the clusters.
2. Resolution Phase:
   • Filtering: Apply filtering techniques to reduce the domains of variables.
   • Search: Perform the search algorithm guided by the evaluation tree and variable clusters.
3. Post-Processing Phase:
   • Tradeoff Reasoning: Integrate user preferences and additional criteria to select the best solution from the set of acceptable solutions.

The system’s user interface provides visualizations of the constraint network, variable clusters, and the evaluation tree, allowing users to follow the resolution process step by step. This transparency aids users in understanding how setpoints are determined and provides insights into the underlying reasoning.

E. Implementation

The implementation of SERRISTE involves several technical aspects:

1. Programming Language: The choice of programming language is crucial for handling numerical computations and managing the resolution process. SERRISTE is implemented in a language that supports efficient numerical operations and complex data structures.
2. Data Structures: Efficient data structures are used to represent variables, domains, constraints, clusters, and the evaluation tree. These structures facilitate quick access and manipulation of information during the resolution process.
3. Algorithm Optimization: The filtering and search algorithms are optimized for performance. Techniques such as heuristic-guided search, intelligent backtracking, and parallel processing (if applicable) are employed to enhance efficiency.
4. User Interface: A graphical user interface (GUI) is developed to provide users with interactive visualizations of the constraint network, clusters, and the evaluation tree. The GUI also allows users to input preferences and criteria for tradeoff reasoning.

In conclusion, SERRISTE leverages artificial intelligence techniques, particularly the constraint satisfaction problem framework, to determine optimal setpoints for greenhouse climate control. By integrating knowledge representation, variable clustering, domain filtering, and systematic search, the system provides a robust and efficient solution to this complex problem.

The SERRISTE system, developed using KAPPA, an object-oriented hybrid environment, incorporates multiple features such as objects, rules, functions, and conventional programming techniques. This system operates under the MS-WINDOWS and MS-DOS operating systems and integrates with the C programming language. KAPPA’s objects are structured hierarchically, with properties defined in slots that include restrictions and behaviors defined by methods. Demons can be attached to slots to automatically respond to alterations. Forward and backward reasoning tasks are performed using rules.

Key Classes in SERRISTE:

1. Quantités: Handles variables and parameters.
2. Contraintes: Manages different types of constraints based on the number of variables.
3. Solutions: Stores acceptable solutions found by the system.
   • Solutions-Acceptables: Instances of acceptable solutions.
   • Solutions-Preferees: Instance of the preferred solution.
4. Groupes: Manages clusters of variables involved in specific applications.

An important object in the system is Resolution, which contains top-level methods for the resolution process. Real-world elements relevant to climate management are represented through objects like Serre (greenhouse), Culture (crop), and Temps (time). Weather forecasts and greenhouse climate measurements are represented by instances of Meteos and ClimatsMesures respectively, named according to the date of consideration (e.g., MeteoJ, ClimatJ).

Example Instances

1. CTsol (Instance of Variables Subclass)
   • Represents soil temperature setpoint variable.
   • Slots include values for different resolution stages.
   • No local methods.
2. C10 (Instance of ContraintesBinaires Subclass)
   • Represents the constraint that the difference between diurnal and nocturnal average temperatures must be within a specified range.
   • Local method Determiner_a3Max determines the value of a3Max.

Rules and Knowledge Base

SERRISTE uses forward chaining rules for deducing parameter values based on input data. For instance, the rule DCTsolAdapto sets the value of DCTsolAdapt based on conditions like crop health, vigor, and weather data.

The knowledge base of SERRISTE includes:

• 24 variables
• 43 parameters
• 27 constraints
• 62 rules

Discussion

SERRISTE aims to enhance greenhouse climate management by integrating heuristic knowledge with AI techniques. It uses constraints on numerical variables for decision support, differentiating itself from conventional expert systems. Though still a prototype, SERRISTE shows promise in improving crop management by leveraging weather predictions and qualitative data like disease symptoms.

Current Status and Future Work

SERRISTE’s decision-making capabilities can prevent diseases and maintain coherence in crop management. The system has been tested for programming errors and is undergoing validation. Preliminary experiments indicate positive results, but further testing and development are needed.

Future Directions:

• Extend representation capabilities to handle soft constraints using fuzzy set theory.
• Address greenhouse management issues for higher outside temperatures.
• Incorporate management of carbon dioxide enrichment and nutrition.
• Conduct in-depth validation in real greenhouses under various conditions.
• Develop decision support for critical or exceptional situations.
• Explore the exploitation of tomato plants’ temperature integration capabilities for energy saving.

Acknowledgements

The development of SERRISTE was supported by the AFME agency and involved collaboration with INRA colleagues from Bioclimatology and Agrarian Systems and Development Departments.

COMPUTER OPPORTUNITIES IN AGRICULTURE AND HORTICULTURE

W. Day
Process Engineering Division
Silsoe Research Institute
Wrest Park, Silsoe, Bedford, UK

I. INTRODUCTION

The revolution in computing technology is significantly impacting process optimization and control across various industries, including agriculture and horticulture. Many technological advancements in computing developed for other industries, like automotive, are being adopted in agriculture due to the larger market size supporting extensive research and development. Nevertheless, agriculture presents unique challenges suitable for the application of new technologies. This chapter discusses the critical themes that will drive the advancement of computer applications in agriculture and horticulture.

With rapid advancements in processing power, compactness, and reduced unit costs, new applications in agriculture and horticulture are becoming feasible. These applications reflect the wide variety of agricultural processes and bring specific technological challenges, including biological variability, complex physiological interactions, and environmental unpredictability. Progress in hardware and software is enabling practical implementations in areas like produce grading, crop yield prediction, and management of crops and livestock.

II. IMAGE ANALYSIS

In horticulture and agriculture, many tasks require human decision-making based on visual information, such as produce harvesting and grading, assessing animal health, and identifying plant diseases. Computer-based image analysis can tackle these tasks, despite challenges posed by biological variability and unstructured environments.

Technological Advances:

• TV Camera Technology: Standard TV cameras can produce images with 512 x 512 pixels and 8-bit intensity levels, adequate for many agricultural scenes. The low costs of this technology, developed for the mass TV market, make it feasible for agricultural applications.
• Processing Speed: Advances in processing speed, currently estimated to be doubling every three years, allow the rapid handling of large amounts of image data.

Practical Applications:

• Produce Grading: Image analysis is already being used for repetitive tasks such as produce grading. Research is ongoing to tackle more complex tasks and develop generic approaches to 3D structural interpretation and tracking of moving images.

Advanced Image Interpretation:

• Artificial Intelligence: AI, particularly intelligent knowledge-based systems, can utilize past knowledge for current tasks. Systems use an ‘inference engine’ for reasoning.
• Neural Networks: These allow computers to learn from examples, useful for tasks where quality attributes are easy for humans to identify but hard to quantify. For instance, neural networks have been used in the classification of apples and identification of chrysanthemum nodes.

Image Analysis for Robot Control:

• Robotic Handling: Image analysis is crucial for robotic tasks in agriculture, such as mushroom harvesting and processing geranium cuttings, where precise control and rapid processing are needed.

III. CROP MODELS AND OPTIMIZATION

Computers have been used extensively in agricultural research to analyze and predict crop growth and responses to environmental factors.

Field Crop Models:

• Real-Time Prediction: Models of field crop performance are being used in advisory roles, such as fertilization timing, although complete models for real-time prediction are challenging due to complex interactions.
• Environmental and Policy Evaluations: Models can predict the impact of climate change and evaluate agricultural policies, requiring robust validation to ensure accuracy.

Greenhouse Crop Models:

• Environmental Control: Greenhouse conditions simplify modeling, allowing real-time control of crop management. Models help optimize variables like CO2 concentration, balancing crop value against the cost of CO2 supply.

The integration of advanced computing technologies in agriculture and horticulture holds significant promise for enhancing productivity and sustainability. Continued advancements in image analysis, AI, and crop modeling will drive the development of practical and efficient solutions for the agricultural sector.

The excerpt discusses the integration of information technology (IT) in agriculture, particularly focusing on greenhouse management and selective field operations. Here’s a summary of the key points:

1. Complexity of Greenhouse Management

• Optimal control in greenhouses involves managing various factors such as internal and external climate conditions, weather forecasts, market prices, and grower preferences.
• Achieving this requires comprehensive information systems to gather and analyze data from multiple sources.

2. Role of Information Technology in Farming

• IT is revolutionizing various industries, including agriculture, but its full potential in farming is just starting to be realized.
• IT farms can integrate various management information systems to aid in both strategic and tactical decision-making.

3. Current Applications of IT

• Greenhouse control systems demonstrate IT’s ability to optimize heating and ventilation in real-time.
• In field agriculture, selective or spatially variable operations are emerging. This involves tailoring farming practices based on variations in soil type, slope, water table height, and weed problems within a field.

4. Selective Field Operations

• By adapting operations to specific parts of a field, farmers can enhance economic and environmental efficiency. This includes optimizing soil conditions and reducing agrochemical usage.
• Technologies enabling selective operations need to assess key parameters across the field, integrating current data with historical performance.

5. Examples of Selective Operations

• Patch Spraying: Research indicates some weeds grow in patches, allowing for targeted herbicide application, reducing overall chemical use and costs.
• This requires advanced computer technologies, including image analysis for real-time weed identification and satellite positioning for accurate sprayer location.

6. Expert Systems in Agriculture

• Expert systems serve as a bridge between research and practical application, providing farmers with the necessary knowledge to make informed decisions.
• They use heuristic problem-solving, can adapt to new information, and combine qualitative and quantitative analyses to enhance decision-making.

7. Addressing Environmental Challenges

• The need to manage animal waste and meet regulatory constraints highlights the importance of integrating expert systems in agriculture. These systems can help farmers navigate environmental issues while optimizing waste management practices.

Conclusion

The effective use of information technology in agriculture, particularly through integrated systems and expert decision-making tools, promises to enhance efficiency, reduce costs, and meet environmental demands. The integration of these technologies will continue to evolve, offering new solutions for both greenhouse management and field operations.

About Us

Welcome to Agriculture Novel, your go-to source for in-depth information and insights into the world of agriculture, hydroponics, and sustainable farming. Our mission is to educate, inspire, and empower a new generation of farmers, hobbyists, and eco-conscious enthusiasts. Whether you’re interested in traditional farming practices or modern innovations, we aim to provide comprehensive guides, expert tips, and the latest updates in agriculture and urban farming.

At Agriculture Novel, we believe in the power of knowledge to transform the way we grow, sustain, and nourish our world. Explore our articles on topics like Fruit Growing Guide, Hydroponics,  Plant Deficiency Guide, and more.

Thank you for joining us on this journey towards a greener, more sustainable future!


About Agronique Horizon
At Agronique Horizon, we specialize in delivering comprehensive digital marketing and web development solutions tailored for the agriculture and hydroponics industries. From custom website design and app development to social media management, we provide end-to-end support for brands aiming to make a meaningful impact. Our team also offers innovative solutions for the real estate sector, bringing precision and visibility to your projects. Learn more about our services here and discover how we can elevate your digital presence