PhD Internship Project: DC2: Adaptive sensor- and context-aware learning

The candidate will be hosted by the CTO organization of NXP Semiconductors at the High Tech Campus in Eindhoven, where you can collaborate with researchers working on tinyML machine learning, embedded systems and Pixelated Capacitive Sensor (PCS) technology. In parallel the candidate will be registered at the faculty of Electrical Engineering, Mathematics and Computer Science of Delft University of Technology for the PhD graduation. This requires taking courses to collect the credits required for the PhD graduation. Some courses will be provided by the Marie Skłodowska-Curie Actions - Doctoral Networks (DN) project, ANT.

Mobility Rule - MSCA-DN

To be eligible to participate in the program, researchers must not have resided or carried out their main activity (work, studies, etc.) in the country of the recruiting beneficiary for more than 12 months in the 36 months immediately before their date of recruitment. Compulsory national service, short stays such as holidays, and time spent as part of a procedure for obtaining refugee status under the Geneva Convention1 are not taken into account. For international European research organisations , international organisations, or entities created under Union law, recruited researchers must not have spent more than 12 months in the 36 months immediately before their date of recruitment in the same appointing organisation. Note that the mobility rule applies to the (first) beneficiary where the researcher is recruited. In case of multiple recruitments, the mobility rule only applies to the first recruitment.

A detailed guide on the eligibility criteria for the applicants of the Marie Skłodowska-Curie Actions - Doctoral Networks (DN) program can be found through this link.

Objectives

To design, build and test a disruptive Embedded AI system with a combination of 1) a novel MCX-N microcontroller of NXP Semiconductors with embedded ML, math, DSP and crypto accelerators, 2) a wireless network link module, and 3) a microcontroller of NXP with embedded CMOS Pixelated Capacitive Sensor (PCS) physical multi-sensing interface.

Expected Results

Embedded AI boards using novel ML-enabled MCX-N microcontrollers and PCS multi-sensing Highly compressed AI models for inference and on-device learning Live demo and field-testing.

PhD enrolment: Doctoral School of TU Delft

Planned secondments

• TU Darmstadt (3 months, M16-M18): Context-aware embedded AI boards for zero-power embedded devices, with Prof. M. Zimmerling (KPI: joint paper)
• University of Trento (4 months, M26-M29): Highly compressed models for zero-power embedded devices (KPI: joint paper)

Candidate profile: electrical engineering, embedded systems, computer science

Desirable skills/interests: embedded systems, machine learning, tinyML, AIML models, building live demonstrators from sensors, microcontrollers, radio modules, etc.

Creating Secure Connections and Infrastructure for a Smarter World

NXP Semiconductors N.V. (NASDAQ: NXPI) makes products and environments safer, more sustainable, and more secure with innovative connectivity and edge processing solutions for a smarter world.We are in the business of better. Not just better technologies, but better innovations to improve society. As the world leader in secure connectivity and processing solutions for embedded applications, NXP is solving the world’s most complex technology challenges to accelerate business innovation, enhance how we work, and advance how we live.

More information about NXP in the Netherlands...