Low-power microcontroller implementation of machine learning models for smart-agriculture applications
Research Groups MiNES (Micro&Nano Electronic Systems)
Thesis type EXPERIMENTAL
Description The candidate will be responsible for implementing pre-trained machine learning models based on neural networks on a microcontroller, helpful in evaluating the health status of a plant starting from impedance and environmental parameters. In particular, it will be necessary to manage the phase of acquiring the required data through existing sensors and interfacing it with the predictive algorithm to be implemented. The ultimate goal is to obtain an effective and energetically autonomous monitoring system capable of operating on a low-cost platform such as a microcontroller. For this reason, particular attention must be paid to the analysis of the resources necessary for executing the algorithm as well as the minimization of the power consumption.
Required skills Microcontroller programming (C)
Deadline 21/11/2023 PROPONI LA TUA CANDIDATURA