Pavement type identification method based on PVDF piezoelectric film and wheel cover near field radar multi-source fusion

Abstract

The invention discloses a pavement type identification method based on PVDF piezoelectric film and wheel cover near-field radar multi-source fusion, and the method comprises the steps: embedding a PVDF piezoelectric film array in a tire, and collecting a dynamic strain signal of a contact spot; a near-field radar is arranged in the wheel cover to detect front road surface echoes; based on the vehicle speed, PVDF time signals are resampled into a unified spatial scale, a radar ground area is dynamically determined in combination with the vehicle body posture, and multi-modal data space alignment is achieved; and respectively extracting PVDF time-frequency characteristics, radar energy distribution, attenuation slope and mirror reflection characteristics, inputting the PVDF time-frequency characteristics, the radar energy distribution, the attenuation slope and the mirror reflection characteristics into a Transform model for cross-modal fusion, and outputting a pavement type and confidence. Through a combined mechanism of radar preview and PVDF tactile verification, the recognition robustness and real-time performance on complex road surfaces such as water accumulation and ice and snow are remarkably improved, and the method is suitable for stability control of high-grade automatic driving vehicles.

Images