Sensor and Odor Display Technologies

Tampere University works on a self-learning odor display, which enables the transmission and synthesis of body odors across time and space. SmartNanotubes Technologies provides their sensor platform, featuring the newly developed receptor molecules for sensing body odors.

Development of a Self-Learning Multichannel Odor Display (Work Package 6)

BACKGROUND: A key challenge in digital olfaction is odor synthesis, which refers to the reproduction of odors for human and machine olfaction using a subset of chemical compounds identified from an original odor source. In terms of producing scents, several devices have been suggested as olfactory displays (ODs). ODs generate odors by accurate mixing and releasing. They can be wearable or placed in the ambient environment. Even though numerous ODs have been developed, they all suffer from the lack of precise mixing and insufficient spatial, temporal, and volumetric control. Thus, improved OD technology is needed.

METHODS: In Work Package 6, Tampere University will develop a self-learning odor display prototype. The aim is to study if a recipe for an unknown scent can be produced from known odor components using an electronic nose (eNose), an olfactory display, and optimization algorithms. Using the eNose, the odor display first measures the electronic fingerprint of the target odor. The first approximation of the target odor is reproduced by the olfactory display. The reproduced odor is measured again by the eNose and the two fingerprints are compared. Certain optimization methods (i.e. stochastic search algorithms) are used to iteratively improve the recipe of the reproduced odor to minimize the difference between the two responses. Human tests are also used to verify that the synthesized odor has high resemblance to the original one. The display can act as a terminal device for delivering odors transferred over the communication network.

Data Integration and Software Development (Work Package 7)

In Work Package 7, SmartNanotubes Technologies will integrate information on healthy and pathological body odors, their description and perception by humans, and their chemical composition with sensor fingerprint for samples of body odors from healthy / disease body odors. The goal is to make a functional software architecture for these applications.

METHODS: All partners will be provided with the eNose devices Smell Inspector with the current version of its software Smell Annotator. Measurements will be carried out by the partners in the other work packages and added to create a body odor smell database. After the development of mucin-derived functionalization in Work Package 3 and the production of corresponding detectors, the next data collection will be based on the newly produced detectors and the performance of the new sensors will be evaluated and optimized.

Additionally, an interface to the software of the odor display (Work Package 6) will be developed. Body odor recognition based on Smell Inspector device, new detectors with mucin-derived smell-receptor functionalization, and an extension of Smell Annotator software will be demonstrated. Our proof-of-principle demonstrators will eventually enable the world’s first digital transmission of a body odor profile between Dresden and Tampere.