The proposed technique computes the roto-translation parameters of multiple mmWave radar devices respect to a reference coordinate system. The method calibrates radars in pairs and it is based on computing the optimal rigid transformation that matches the movement trajectories of multiple targets in the environment that are tracked by both radars. A dedicated association cost function has been designed that is able to identify the best track-to-track matches to perform calibration. The cost function is composed of terms that account for (i) the residual error of two tracks after application of roto-translation, (ii) the time synchronization between two tracks, (iii) the length of tracks. The technique is then able to automatically evaluate quality of the calibration by computing the residual matching error and re-calibrate if needed.

TRL (Technology Readiness Level)

TRL 4 / 9

Inventors

Michele
Rossi

Jacopo
Pegoraro

Marco
Canil

Anish
Shastri

Paolo
Casari

Download the brochure

  • MERCURIO – Communication system based on multipole-phase division

    The patent discloses a spatial division multiplexing optical communication system comprising optical devices for the multiplexing and demultiplexing of beams endowed with different multipole phases. This invention proposes an innovative and efficient solution to increase the transmission capacity of the[...]

  • A Method and system for reconstructing the micro-Doppler spectrogram from estimated scattered channel in communication and sensing joint applications

    There is growing interest in human tracking, human activity recognition (HAR), and person identification by reusing wireless signals transmitted by existing communication devices, such as WiFi and cellular radio transceivers. Radiofrequency sensing systems have emerged as viable alternatives to cameras[...]

  • Device authentication through earphones

    This patent covers a method than can authenticate a person using earphones, thanks to the movement of the mouth. The earphones record the shape of the ear canal during a predefined movement of the mandible and a machine learning algorithm[...]

Subscribe to newsletter

Follow the activities promoted by UniSMART Foundation