Passenger detection in cars with small form-factor IR sensors (Grid-eye)

In this paper, an experiment of passenger detection in cars is presented. A small and cost-effective, commercial IR sensor was used to investigate the possibility of detecting people in the different seats of the car. The electronics are based on the Arduino platform, which is used to communicate with a PC to collect the data, and an AMG8833 IR sensor board, which enables fast and efficient temperature detection in 8×8 matrix. The Grid-Eye sensor has its limits, so these aspects were also investigated, according to the positioning, the resolution and temperature differences, which parameters point to the maximum number of detectable people. The sensor was validated in room temperature laboratory environment, where it was possible to differentiate a person from the background. Four positions were considered in the car, with front, and back seated people (0-5 persons) for different use cases (Case A-D). It was found, that with the aforementioned limitations, at given positions, it was possible to differentiate seated and non-seated positions in the car, thus with more sensor nodes, the passenger detection is possible with a low-cost solution.