Real-Time Edge Computing for Road Surface Classification Using Multi-IMU Data and a Hybrid CNN-LSTM Classification Model
Luis A. Arce-Saenz, Luis A. Salazar-Calderón, Renato Galluzzi, Javier Izquierdo-Reyes, Rogelio Bustamante-Bello · 2026
Read Real-Time Edge Computing for Road Surface Classification Using Multi-IMU Data and a Hybrid CNN-LSTM Classification Model by Luis A. Arce-Saenz, Luis A. Salazar-Calderón, Renato Galluzzi, Javier Izquierdo-Reyes, Rogelio Bustamante-Bello (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13364251.txt.
Description
Luis A Arce-Saenz and colleagues present a 2026 study on classifying road surfaces in real time from vibration sensors inside a moving car. Their hybrid network, part convolutional and part memory-based, reads streams from several inertial units and sorts the road into five conditions. It runs on an embedded automotive computer, and the team tuned the deployment for speed at city driving pace.
The edge version answered in about four milliseconds, a cut of nearly nine tenths versus the offline setup, inside the budget set by the sensor rate. Its score dipped modestly versus the offline model but stayed fair, and mapping pinned surface faults to places regardless of light, a known weakness of camera methods. Highway speeds are the next test.
Catalog record
| CREATOR | Luis A. Arce-Saenz, Luis A. Salazar-Calderón, Renato Galluzzi, Javier Izquierdo-Reyes, Rogelio Bustamante-Bello |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 48974 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |