Michael Madsen


Extractions from my LinkedIn and ResearchGate profiles:





Open-Source Do-It-Yourself machine olfaction. That is what I do.

Machine olfaction (i.e. artificial perception of "smell", electronic nose or just e-nose for short) is a relatively new field, which has come of age as a full-grown technical science with recent advances in diverse disciplines such as nano science and chemometrics. In short, an e-nose mimics the sense of smell by use of advanced sensor technology, refined sampling methodologies, and multivariate data analysis.

My ambition is to further develop my current e-nose prototypes into commercial products. The simplest 8 ch devices are suitable for educational purposes, the 16 ch devices are adequate for the process engineer, and the 32 ch devices are intended for the ambitious scientist keen to dissect the challenging World of olfaction.

The workhorses are classical Taguchi gas sensors (sintered SnO2 doped with different metals) and machine learning algorithms. Simple, cheap, robust, and indeed quite capable.

The first e-nose systems have been deployed and are currently being tested by a dedicated user group. End-user involvement in this early phase is of paramount importance as it helps me to guide further development efforts, while optimising system design and refining user protocols.

As of September 2016 I am in the prototyping stage looking for academic and industrial partners as well as seeding/start-up capital.

If you are interested in a collaboration, please do not hesitate to contact me. I am always open to making new professional acquaintances.

For the technically interested, documentation is gradually building up and is put online on my private website.

Elaborate e-nose project description: www.maskau.dk/projects/electronic-nose

Open source hardware designs: www.maskau.dk/embedded

Some literature on the topic machine olfaction: www.mendeley.com/groups/2952311/electronic-noses