In-Vehicle Occupant Detection Based on Zone Source Feature Using MM-Wave MIMO Radar
Huacong Tan, Zhaocheng Yang · 2021 CIE International Conference on Radar (Radar) · 2021
Millimeter wave (mm-wave) radar is generally used as an autonomous driving sensor for road environment detection, but it can also play an important role in vehicle interior sensing. In this paper, we propose an in-vehicle occupant detection method based on zone source feature using mm-wave multiple-input multiple-output (MIMO) radar. The zone source feature is extracted from super-resolution range-azimuth (RA) map constructed by multiple signal classification (MUSIC) algorithm. Whether the seat is in absence or occupancy status, the zone source feature basically follows normal distribution and have very little overlap according to the experiment data. Therefore, low constant false alarm rate (1%) can be obtained by setting an appropriate threshold while achieving high detection rate (99.09%). Moreover, the proposed method is implemented on embedded digital signal processor (DSP) and verified in real-time stationary and driving experiment. The high accuracies indicate its robustness and practicability.