Student Project: Developing a data-driven algorithm to detect plant health and irrigation status

Develop a data-driven technique to turn a fine-grained time series on the electrical currents of the hydroponic solution at the return drain in a cucumber greenhouse into valuable information for horticulture growers.

What you will do

Within our Autonomous Greenhouse program, we conduct research to develop a greenhouse that can operate fully without human intervention. An important parameter herein is the nutrient concentration of the irrigation solution that is used to feed the crop. This concentration is estimated from the electroconductivity (EC) of the irrigation and drain water. Imec has developed a compact and low-power EC sensor, which we have deployed in a greenhouse compartment to continuously monitor the performance of individual plants. This gives a large time-series dataset, from which we want to extract actionable information that helps to control the greenhouse.

Below a set of projected use cases for this dataset:

• Visualization and assessment of health and performance conditions of individual plants.
• Detection of defects and anomalies in the irrigation system.
• Explore and validate links between sensed conductivities and plant health and irrigation status.
• Optimization of the day-to-day irrigation strategy.

The internship work and activities will be organized with a scrum-like methodology. Prioritized tasks will be selected from the backlog and will be tackled and evaluated on a biweekly basis. At the end of each biweekly iteration, you will showcase the progress made and will reflect on gained insights and possible improvements to focus on. Additional stakeholders or users may make part of the showcases to get better feedback on the developed product. An initial backlog for the internship will be built based on the projected uses mentioned above.

Main tasks:

• Data pre-processing and feature extraction based on domain knowledge, literature and exploratory data analysis.
• Creating data visualizations to share with stakeholders (internal and external) and end users.
• Use machine learning and time series analysis techniques for pattern detection and classification to understand the irrigation-plant response cycle.
• Explore ways to use the developed ML (or relevant) techniques to support growers’ irrigation decisions.
• Collaborate and brainstorm with other team members and experts, provide regular update on status and results.
• Visit greenhouse, interact with growers in order to identify pain points and other use cases.

What we do for you

• We have a challenging problem where you have freedom to explore and deliver solutions.
• We have a diverse team of experts in data science, machine learning, hardware, software, sensors, biology, and agriculture who can coach you and provide you with advice in the development of the assignment.
• You will join the Digital Twin team of OnePlanet, which employs state of the art knowledge on machine learning for precision agriculture and the frameworks necessary to perform these big data tasks at large scale.
• You will be able to exchange views and knowledge with the OnePlanet and Imec community of experts and scientists, widening your professional network.
• We can help you improve your coding skills up to industry standards.
• You will have access to our cloud infrastructure to solve this problem allowing you to process large amount of data within reasonable time.

Who you are

• You are MSc student in Data Science, Applied Informatics, Farm Technology, or related fields.
• You are available for a period of 4 - 6 months.
• Knowledge of Python.
• (Advanced) Knowledge of supervised and unsupervised machine learning algorithms, testing and validation methods.
• Knowledge of time series analysis and anomaly detection.
• (Advanced) Experience with Numpy, Pandas, Matplotlib, Scikit-Learn.
• Experience with PyTorch or TensorFlow is a plus.
• Knowledge of Git.
• Basic understanding of Agile-Scrum.
• You are entitled to do an internship in the Netherlands.
• You are self-starter and able to work independently.
• Good written and verbal English skills.

Interested

Does this project sound like an interesting next step in your career at imec? Don’t hesitate to submit your application by clicking on ‘APPLY NOW’.
Should you have more questions about the project and the recruitment process? Martijn Kohl of the Talent Acquisition Team will be happy to assist you.