PORTALE DELLA DIDATTICA

Ricerca CERCA
  KEYWORD

Techniques for sensor drift compensation for micro neural processor implementation

keywords DRIFT COMPENSATION, MACHINE LEARNING, ARTIFICIAL NEURAL NETWORKS, MICRO NEURAL PROCESSOR, SENSOR DRIFT

Reference persons LUCIANO LAVAGNO, MIHAI TEODOR LAZARESCU

External reference persons Ing. Roberto Simmarano dell'azienda Sensichips s.r.l.

Research Groups Microelectronics, Sensing and processing

Thesis type COMPANY STAGE

Description Sensor readings can drift over time as a result of chemical or physical contamination and wear. These drifts impact the sensor's capability to detect the target analyte with precision and accuracy.

The thesis aims to review existing literature on drift mitigation methods and analyze the causes and effects of drifts on the readings of the company's sensors. Next, will be analyzed and tested machine learning and other techniques to correct sensor drifts using offline sensor data. Algorithm analysis and selection will consider the limited resources available from the company's proprietary micro neural processor.

The thesis has a high impact as its objective is to enhance the lifespan and functionality of sensors with techniques applicable to various types of sensors.

Objectives: Improve sensor capability to detect the target analyte with precision and accuracy by on-site corrective techniques to mitigate sensor drift, as well as evaluating their compatibility with the constraints of the company's micro neural processor.

See also  http://www.sensichips.com/

Required skills No specific skills are required.

Useful skills include Sensor technology (principles and applications, causes and effects of sensor drift), data analysis (statistical techniques, identify drift patterns and trends, evaluate accuracy and precision), machine learning basics, microelectronics (microprocessor architectures).

Notes The thesis will be carried out in collaboration with Sensichips s.r.l., specialized in the development of highly miniaturized learning microsensors. Its sensors are based on an innovative microelectronic multi-sensor chip platform code-named SENIPLUS.


Deadline 20/11/2024      PROPONI LA TUA CANDIDATURA