Artificial intelligence in viticulture and wine research

Using artificial intelligence to obtain an objective quality assessment of wine


The aim of the PINOT project is to study and implement sensor-based AI systems along the entire wine value chain, from the grape to the consumer.

The acquired sensor data facilitates the work of winegrowers and cellar masters in terms of production and quality assurance by digitalizing and documenting the aroma, taste and texture of wine, thereby making these parameters reproducible.

Subjecting wine to multisensory analysis enables dealers and sommeliers to pretest it, check that delivery matches the order, and offer the wine to specific target groups, while customers benefit by receiving the wine that best suits their palate.


The art of winemaking is highly complex and calls for a great deal of expertise and skill. Numerous factors affect wine quality, such as climatic conditions, soil properties, and the individual processing steps taken by the winegrower.

Various tests are carried out to determine a wine’s authenticity and quality, from physical and chemical analyses to tasting sessions with professional tasters. In the latter case particularly, analysis and documentation may vary significantly as they depend on the individual’s sense of taste. Moreover, wine tasting and analyses are both extremely and time-consuming.

This is precisely where the PINOT project comes in. Multisensor systems and artificial intelligence enable the taste of wine to be captured holistically and objectively, while digitalizing its sensory perception.

Digitalizing human sensory perception

Near-sensor AI facilitates the quality assurance of wine

KI im Weinbau

To digitalize the human sense of smell and taste, we at Fraunhofer IIS are developing specific sensor technology, which we

  • evaluate, analyze and consolidate using different AI methods,
  • implement in suitable prototype systems, and
  • test under controlled and real conditions to achieve the greatest possible sensor selectivity, sensitivity and specificity.



Weincampus Neustadt in cooperation with Dienstleistungszentrum Ländlicher Raum (DLR) Rheinpfalz

  • Project coordination and management
  • Collection and provision of wine chemistry and human sensory data

Genie Enterprise Inc.

  • Requirements analysis (specification, documentation, organization)
  • Selection, implementation and training of appropriate AI models for sensor data fusion and the human-AI interface

Wille Engineering

  • Conceptual design and development of various hardware prototypes (memory, power consumption, etc.)

Vineyard Cloud GmbH

  • Project application (development of suitable business models, market analyses)

Fraunhofer IIS

  • AI-based selection, evaluation and validation of appropriate (gas) sensor technology
  • Feasibility study on analyzing wine aromas using gas sensor technology
  • Functionality monitoring of the sensor systems

Trier University of Applied Sciences, Environmental Campus Birkenfeld

  • Further development of a smart software platform to process training data and provide AI algorithms

You may also read:


AI-based Software for Scent Prediction


Parfum/Aroma Ingredient Discovery | Materials Design


Smart Sensors for Gas and Odor Detection

We offer system solutions for application-specific gas sensors and their AI-based evaluation.


Sensory Perception


Why we are the perfect partner:

  • State-of-the-art AI performance
    by applying proprietary machine learning methods for scent prediction
  • Experience in application-specific development of AI solutions
    through many successful development projects
  • Many years of experience as a partner in large industrial projects
    from the development of the first idea up to consumer goods