1. What limitations of existing occupancy sensing technologies motivated this work?
Air bag systems have traditionally protected large, white males the most. If you are a small female, teenager or slighter frame the airbags can actually harm you more than they protect you. The airbag technology is now getting more sophisticated, so they can partially deploy or do more targeted deployment locations (e.g. short and wide, versus long and narrow). However if you do not know more information about the person who is sitting in the seat, and what position they are in then you can not use those features.
The traditional POS or weight sensors are just not accurate enough.
For the last few years the industry turned to cameras and radars to detect occupants. Both systems have limitations. A driver focused camera is in the car to stay. It works well, but using that single camera to monitor all occupants has been shown to have limitations. For example when the driver is very large and blocks the line of sight to the 2nd row of seats. Radars were also thought to provide answers, but they produce far too many false positives and negatives, and are proving to be too difficult and expensive to integrate. In a recent OCS study run by BMW our technology was up against 3 Tier1 solutions using mainly camera, radar or existing seat sensors. We came out on top.
2. How does tactile sensing improve classification accuracy?
We have proven ourselves to be a new, more innovative, and higher accurate, solution for Occupancy classification. One that really understands the occupant, detect their pose and position, and even detects how comfortable they are feeling.
Our tactile seat sensors are providing a whole data set that has never existed before in the car. A true occupant monitoring system from the seat perspective. With an ability to detect movements and adjustments that cameras and radar do not notice. At a detection and reaction speed that provides true realtime position and movement data. The processing power, and thus cost, of capturing the data is very low, making the dataset one of the most affordable available to OEMs.
We also believe deeply in sensor fusion. Double verifying using different sensors technology to increase robustness, and consumer confidence. We design our sensors with scalability and integration in mind. Ensuring simplicity of integration and richness of data at every step of the process. We are designing everything we do with the aim of having these sensors in every car.
3. Which occupant scenarios are most difficult to classify reliably?
Traditionally seat sensors are very good at classifying large adults. Historically technology has struggled to differentiate children from small adults. A 6 year old child weighs about 20kgs, where as a 5th percentile female is about 50kgs.
At the moment the regulations ask car companies to be able to distinguish between these two use cases. That leave a very high percentage of larger children and teenagers in an “undefined grey zone” where cars are not really able to protect them. We think that is unacceptable. Even regulatory bodies like the FMVSS have stated they would like to improve this, and the technology companies have argued it is not possible. We believe with our technology it now is.
4. How might tactile sensing influence future restraint systems?
We believe that everyone who is in a car should be protected. So we foresee a future where the restraint systems react very differently depending on the occupant.
We believe that we will be able to sense presence as well as touch, and that will be able to improve current systems like “hands-on wheel” or “air-bag suppression zones” where the technology sometimes annoys occupants more than they protect. It will be important that technology works with the occupants and that the occupants trust the signals and alarms that they hear.
We believe the technology will not only help make cars more safe, but also more comfortable. Helping the OEMs understand driving dynamics and develop more comfort and personalization features as a result. As the driving experience evolves to be more of a travelling experience, brand loyalty may be based more on how you feel in the car rather than how you feel driving the car.
Interested in exterior vehicle sensing technology?
With a pass to InCabin Europe, you’ll also get full access to our
co-located sister event, AutoSens. Find out more here >>
Watch more top sessions and technical interviews on the InCabin YouTube channel ⬇
