“Automotive designers have successfully integrated millimeter wave (mmWave) sensors into multiple automotive cabin applications. One of these applications is the ability to detect occupants in the car in various lighting conditions and sensor placement, regardless of whether they are moving. This helps the car system to detect the location of children or people left unattended in the car for temperature control.
Author: Texas Instruments Alessandro Veglio
Automotive designers have successfully integrated millimeter wave (mmWave) sensors into multiple automotive cabin applications.
One of these applications is the ability to detect occupants in the car in various lighting conditions and sensor placement, regardless of whether they are moving. This helps the car system to detect the location of children or people left unattended in the car for temperature control.
Azcom Technology demonstrated how the AWR1642 millimeter wave sensor combined with Azcom’s proprietary algorithm can reliably identify the seating situation of a person on the seat. We drove at different speeds, different environments (city, highway) and cab (light, temperature) conditions, and analyzed different seat configurations.
In our demonstration, the sensor will be suspended from the sunroof towards the rear seat (as shown in Figure 1), although in the final installation it is more likely to be placed inside the seat back, around the rearview mirror, or inside the roof, etc. . Since the millimeter wave can sense various materials, including the materials that make up the vehicle, the sensing performance will not change when installed in the seat or roof. All processing, including the enhancements of Azcom Technology, runs on the sensor, and the graphical user interface on the host helps to visualize the results.
Figure 1: The millimeter wave sensor installed on the sunroof of the vehicle
The main challenge for this use case is to achieve sufficient detection robustness when the engine is turned on and the car is driving. The combination of these two events introduces a set of interruptions in signals from several vibration patterns that are not present in the static setting. For this reason, we designed a new algorithm. This algorithm is not very sensitive to the vibration from the road, and can detect all possible combinations of people in the car.
In addition to the reference design for passenger seat detection in the car, we have also applied and verified these enhancements. Figures 2, 3, 4, and 5 are some snapshots from the sample drive, as well as diagrams of detecting the seating of people in the car.
In Figure 2, when driving in a city, no one is seated in the rear seat, and the algorithm detects that there is no fault. Statistics are calculated at the ratio of processed frames: 6 fps in this use case. In real products, lower-frequency secondary decision-making tools will make the detection more robust.
PIC 2: No one in the rear seat
In Figure 3, the algorithm successfully detects a person seated in area 1, as shown in the red box.
Figure 3: Seating detected in the rear seat
Figure 4 focuses on the accuracy of the algorithm when using different car cabs. By testing two different models, we demonstrated reliable seat detection when driving at different speeds.
Figure 4: Checking the seating situation of people in different car cabins
Figure 5 shows the expansion of the design to two rows of four seats. Although this scene is more complex and challenging, after special adjustments and optimizations, the algorithm performs as well as the single-row setting.
Figure 5: Four-seater configuration
Relying on its expertise in passenger occupancy detection applications and deep expertise in TI platform, signal processing, RF and embedded system design and development, Azcom Technology provides a series of value-added R&D services that can help you build millimeter wave-enabled products , And greatly shorten the time to market.
To better understand Azcom’s portfolio of design and development services for millimeter wave sensor product development and support, please refer to the Azcom webpage to learn about millimeter waves.