Period, duration
February - July 2023, 20 weeks
Profile required
Master2 or engineering school student
Gratuity
4.05/h (standard)
Location
Annecy-le-vieux ; Maison de la mécatronique ; Symme laboratory
Subject
This internship is part of the development of a device for locating people in a habitat, without the need for wearable sensors or devices such as telephones. Step-sensitive sensors placed in the habitat could locate the footsteps of the inhabitants using the trilateration technique, in order to obtain a "Marauder's Map" type map in Harry Potter. To achieve this, a new type of sensor is used, based on an original, non-Newtonian component: silly putty, combined with graphene.
Initial conclusive tests have led to the development of a sensor demonstrator, whose operating principle is based on the measurement of sudden changes in the conductive properties of a composite material when mechanical stress is applied. Without amplification, the resulting device is as sensitive as the most powerful seismic sensor on the market, which is amplified by a factor of 50.
However, the transducer developed delivers a non-linear response, albeit reproducible for a repeatable load, but with a temporal drift. This behavior makes it difficult to identify the transducer's transfer function using conventional methods. The working hypothesis for continuing the study is that the transducer response can be characterized by a non-linear model, with some parameters undergoing temporal drift. In order to validate this hypothesis, a large number of automated experiments need to be carried out, in particular at low-energy load levels comparable to the micro-seismic energy produced by human footsteps a few meters apart. A test bench comprising a piezoelectric actuator for reproducible vibration generation and an acquisition system are already available in the laboratory.
The scientific challenge lies in processing the data to establish the sensor behavior model. The student recruited will have to set up a test protocol to characterize the responses of the seismic sensor as a function of numerous parameters, and work on the following points :
- measurement automation,
- signal acquisition and data processing,
- sensor learning using artificial intelligence.
At the end of the course, the aim is to determine the transducer's transfer function. We want it to be able to recognize the characteristics (amplitude, emission distance) of a vibration in the case of an unknown load. The candidate will ideally have a profile in instrumentation, with skills in electronics, Labview, computing (data processing, AI) and a strong taste for experimentation. He/she will have access to the SYMME laboratory's equipment (mechatronics room, sensors, 3D printer, electronic components) to carry out his/her work.
Keywords: seismic sensor, automation, characterizations, Labview, data processing.
Contact
Thomas.Mazingue
@univ-smb.fr
Project team
Christine Barthod, teacher-researcher IUT-A / SYMME
Luc Maréchal, teacher-researcher POLYTECH A-C / SYMME
Thomas Mazingue, teacher-researcher POLYTECH A-C / SYMME