[solved] Vehicle Airbag Triggering?

Go To Last Post
13 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Greetings -

 

I hope that General Electronics is a reasonable place for this query .... only other place would be off-topic but I think it is more on-topic than off-topic...

 

I am trying to design an event triggering algorithm for my recording accelerometer. I've been looking at airbag triggering because that seems like the only general application with similar behavior. Hope that someone here on the list has some first hand experience in this area. No, I am not talking about the kind of experience Torby had recently

 

( https://www.avrfreaks.net/forum/...)

 

Anyway, the questions, hoping that they are not too uninformed:

 

1. Is is purely acceleration or is there a rate of change of acceleration involved in the determination (to trigger the airbag)?

 

2. What sort of acceleration trigger levels are involved?

 

3. How does triggering deal with vehicle on a slope along the axis of motion? Is the trigger threshold simply so large compared to the gravity component projected onto that axis through tilt that gravity can just be ignored?

 

4. Any useful references on the subject?

 

Many thanks

Jim

This topic has a solution.

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

Last Edited: Thu. Dec 12, 2019 - 05:20 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Many of the current MEMS accelerometer chips have such "impact" triggering built in.

 

Might be worth looking at some datasheets & associated app notes, etc, to see if they describe how they do it ...

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Just to say that many dashcams also offer this. They keep any recording being made when a MEMS tells them it's an impact situation. I still don't know what the criterion is for "impact" but the point is that you maybe more likely to find an open design for a dashcam than you are to find any documentation for airbags as I imagine the latter is heavily regulated and it's a pretty "closed world" in which such development will be done.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Wikipedia had this to say:

Triggering conditions[edit]

Some cars provide the option to turn off airbags

Airbags are designed to deploy in frontal and near-frontal collisions more severe than a threshold defined by the regulations governing vehicle construction in whatever particular market the vehicle is intended for: United States regulations require deployment in crashes at least equivalent in deceleration to a 23 km/h (14 mph) barrier collision, or similarly, striking a parked car of similar size across the full front of each vehicle at about twice the speed.[102] International regulations are performance based, rather than technology-based, so airbag deployment threshold is a function of overall vehicle design.

Unlike crash tests into barriers, real-world crashes typically occur at angles other than directly into the front of the vehicle, and the crash forces usually are not evenly distributed across the front of the vehicle. Consequently, the relative speed between a striking and struck vehicle required to deploy the airbag in a real-world crash can be much higher than an equivalent barrier crash. Because airbag sensors measure deceleration, vehicle speed is not a good indicator of whether an airbag should have deployed. Airbags can deploy due to the vehicle's undercarriage striking a low object protruding above the roadway due to the resulting deceleration.

The airbag sensor is a MEMS accelerometer, which is a small integrated circuit with integrated micro mechanical elements. The microscopic mechanical element moves in response to rapid deceleration, and this motion causes a change in capacitance, which is detected by the electronics on the chip that then sends a signal to fire the airbag. The most common MEMS accelerometer in use is the ADXL-50 by Analog Devices, but there are other MEMS manufacturers as well.

That correlates to what I have seen in vehicles. Namely - single axis mems accelerometer. Accel/decel is the rate of change of speed,  the derivative of this is distance (??). So I'd say decel is the determinant - accel is useless as airbag deployment would aid the accell. If you get hit from behind with great velocity, the headrest comes into play.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kartman wrote:
Accel/decel is the rate of change of speed,  the derivative of this is distance (??).

Due to the question marks I'll respond.  I've seen the derivative of accel/decel referred to as jerk.  I don't know if this term is used outside of the motor/motion control world.  Distance is the integral of speed.

Letting the smoke out since 1978

 

 

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

digitalDan wrote:
  I've seen the derivative of accel/decel referred to as jerk.  I don't know if this term is used outside of the motor/motion control world.

 

https://en.wikipedia.org/wiki/Jerk_(physics)

 

  Distance is the integral of speed.

Indeed.

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Jerk? I thought that was me.

If you don't know my whole story, keep your mouth shut.

If you know my whole story, you're an accomplice. Keep your mouth shut. 

This reply has been marked as the solution. 
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

With respect to (2) I found this.  Looks like on the order of 30 g's.  I'm guessing you don't have that sort of signal-to-Earth-g ratio.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks for the "discussion" folks. Indeed, velocity is the rate of change (derivative) of position and acceleration is the rate of change (derivative) of velocity. So velocity is the integral of acceleration and position is the integral of velocity. 'Nuf for that.

 

The sensor I am using has a programmable impact detect capability, yes, but like a dumbo, I never routed that signal output to the processor and the leadless chip carrier is impossible to solder a jumper wire to, and I am not going to rev the board at this point. So, if triggering is going to be implemented, it has to be in processor software. I have looked at the description of the sensor algorithm, but, to me, it seems totally wacko. 

 

MattRW - thank you. The big take-away from that reference on this project is that the airbag trigger level appears to be based on acceleration without consideration of jerk, and that the trigger is some 10s of g. This practically means that you do not have to worry about the inclination of the vehicle. For example, San Francisco has a number of street sections with greater than 30% grade. That is a slope of 30m in 100m. Using an online ( https://www.rapidtables.com/calc... ) arctangent calculator, that is an angle of 16 degrees. An accelerometer aligned to the fore-aft axis of a vehicle would report 0.27 g as a consequence of that tilt, even when stationary. Thus, if the impact trigger level is several 10s of g, the practical consequence is pretty small. 

 

So, for practical purposes, THIS specific question has been answered. Of course, there are plenty more in this "project", most of them having to do with Vector Analysis and Numerical Methods. 

 

Many thanks to those who contributed to an interesting discussion.

 

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I think this is a great paper describing air bag triggering.

 

Of note, it has a table (Table 2) of 10 patents for "crash detection", and a description of the parameters used in the determination of a crash, (or not).

It would seem that most use a combination of "Delta V", Accel and Jerk forces.

 

Table 3 lists the Decel (g's) for  several common vehicles, (4.7 - 7.5 g's).

 

This article has an interesting graph that superimposes Speed, Accel, and Jerk vs Time for an off-set head on crash, with calibrated axis labels.

 

JC

Last Edited: Thu. Dec 12, 2019 - 06:41 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

This article has an interesting graph that superimposes Speed, Accel, and Jerk vs Time for an off-set head on crash, with calibrated axis labels.

That's quite interesting, especially figure 3...they work to not really measure the largest components, but to predict them from the initial data (since you want the bag already deployed before the worst of the impact occurs).  Certainly a lot of intellectual property in this area.

 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Apropos of nothing much: if you want to build your own car in the UK, it is illegal to fit an airbag/sensor/trigger - because they're so tightly coupled to the actual vehicle for which they are intended.

 

Neil

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Doc JC -

 

Thanks for that great reference. I do appreciate it.

 

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net