Motion Sensor Algorithm Design “Wide Area Tap” for a wearable device AI approach or Human approach

Société : MOVEA (Groupe TDK)

Founded in 2003, InvenSense Inc., a TDK Group Company, is the world’s leading provider of MEMS sensor platforms. InvenSense’s vision of Sensing Everything™ targets the consumer electronics and industrial markets with integrated Motion and Sound solutions. Our solutions combine MEMS (micro electrical mechanical systems) sensors, such as accelerometers, gyroscopes, compasses, and microphones with proprietary algorithms and firmware that intelligently process, synthesize, and calibrate the output of sensors, maximizing performance and accuracy. InvenSense’s motion tracking, audio and location platforms, and services can be found in many of the world’s largest and most iconic brands including smartphones, tablets, wearables, drones, gaming devices, internet of things, automotive products, and remote controls for smart TVs.

Description du poste

What is it about?
We are looking for motivated students to join our Algo Motion team at TDK/Invensense the world’s leading provider of MEMS sensor platforms (accelerometers, gyroscopes, magnetometers, and microphones). You will join our small and friendly SW team, located in Grenoble city.

TWS (True wireless stereo) earbuds are smarter generations after generations. TDK offers productized software solution to enable high level feature like spatial audio or voice activity detection. As part of the algorithm motion team, the intern will focus on an interface feature named WAT (Wide Area Tap). For WAT, accelerometer sensor is located inside the TWS placed in the user’s ear. When tapping somewhere on his face close to ear but not directly to the device, we expect to detect the gesture from accel signal. Challenge is linked to signal quality due face soft tissues attenuations. Work would be divided in sub-parties described in tasks below.

Job Description
• Learning phase: benchmark existing WAT algorithms and competitors, understand the current algorithm performances and limitations.
• Signal processing phase: understand soft-skin and bone signal conduction, improve the SNR on existing database.
• Design and implementation phase: propose and develop an improved WAT detection algorithm using a new approach (machine learning or state-machine based algorithm).
• Validation of the proposed solution phase: benchmark the proposed solution with the initial TDK one, with performance report. Integrate the solution in an existing demo setup and propose idea for an improved live test demo.

Profil recherché

Duration : 6 months
• Qualifications : BAC + 5 program
• Dates: To be defined
• Location: Grenoble (quartier Europole)

Qualifications
• Machine learning (DNN and other ML methods)
• Programming in Python (or Matlab) and C
• Signal processing is mandatory
• Physics or bio-physics is a plus
• Pugnacity and creativity

Voir le fichier joint