Sensor for Human detection

[rabhishek91]

Hi everyone.

My latest project includes detecting whether the person is seated or not(in automobiles). This can be done by using a pressure sensor or just by a micro switch.

Suppose i'll keep something heavy(Say a desktop pc) on the seat. Then the sensor detects it as a true condition and gives the output. But actually no one(Human being) is seated .
As a solution to this problem i need to detect the presence of human along with the weight .

This can be done by using image processing methods but i don't want to complicate my project . Is there any other sensor or a way which would help me in detecting the presence of human?

 

[alec_t]

Mount a thermistor on/in the seat.

 

[rabhishek91]

Mount a thermistor on/in the seat.

Thanks for the reply. Sir are thermistors that sensitive ?
I mean there may not be significant change of temperature when person is seated. So are thermistors capable of understanding such minor variations

 

[alec_t]

With the right amplification/circuitry even a 1°C change should be reliably detectable. I think the temperature change would be more significant than you imagine (unless ambient temperature for the seat is ~37°C).

 

[rabhishek91]

With the right amplification/circuitry even a 1°C change should be reliably detectable. I think the temperature change would be more significant than you imagine (unless ambient temperature for the seat is ~37°C).

Got it sir. Thank you.
I have a doubt.
I think placing thermistor in the seat won't help because the body temperature of a human may not reach the thermistor through the cushion or seat.
So i think placing thermistor on the seat would be a better option. What your suggestion sir ?

 

[alec_t]

If the seat cover is thin the thermistor could be just below it, in contact with the covering fabric.
Here's a simple circuit using two thermistors; one as the seat sensor and the other as a reference allowed to reach ambient temperature close to the seat squab. Simulation shows that the R1/thermistor junction voltage changes by ~60mV for a 1°C temperature difference between the two thermistors.

 

[rabhishek91]

If the seat cover is thin the thermistor could be just below it, in contact with the covering fabric.
Here's a simple circuit using two thermistors; one as the seat sensor and the other as a reference allowed to reach ambient temperature close to the seat squab. Simulation shows that the R1/thermistor junction voltage changes by ~60mV for a 1°C temperature difference between the two thermistors.
View attachment 72206

Thanks a lot sir.
As you mentioned earlier they have used an amplifier to enhance the signals. I'll work on this circuit and keep you posted.

 

[davenn]

and if the sun is shining through the vehicle window, any thermister in/on the seat is going to false trigger

Dave

 

[rabhishek91]

and if the sun is shining through the vehicle window, any thermistor in/on the seat is going to false trigger

Dave

Yes sir your right.
As a solution to this problem we can use differential amplifier circuit as shown above.
One thermistor tells the surrounding temperature output and other can be used to measure temperature due to presence of human.

 

[ericgibbs]

Yes sir your right.
As a solution to this problem we can use differential amplifier circuit as shown above.
One thermistor tells the surrounding temperature output and other can be used to measure temperature due to presence of human.

hi r91,
The problem would be that the cars interior could be at a higher temperature than the heat transferred thru the humans clothing and car seat to the thermistor.

Your original idea of a mechanical sensor sounds the best solution.

 

[rabhishek91]

hi r91,
The problem would be that the cars interior could be at a higher temperature than the heat transferred thru the humans clothing and car seat to the thermistor.

Your original idea of a mechanical sensor sounds the best solution.

I agree sir. I am thinking of placing IR sensor(preferable 3- to detect occupants irrespective of age) to confirm the presence of human. I have attached a rough sketch below.

 

[Diver300]

A mechanical switch is what is usually used.

I have recorded over 50 °C inside a car that was left in the sun. That was in England where the ambient temperature was less than 30 °C. I guess the seat surface temperature was more like 60 °C, as the black leather seats would have been too hot to sit on in shorts. A temperature sensor would be terrible to detect a person.

 

[jpanhalt]

I am not quite sure what the problem to be addressed here is. Won't the driver know whether the mass in the passenger seat is human or not?

My car gives a warning when there is an object in the front passenger seat and the seat belt is not clipped (something like, "passenger air bag disabled"). It is just a panel light and is easily ignored.

If your car doesn't have that, then a logic circuit that would use the seatbelt switch and the seat switch with a similar warning that the driver could override might suffice.

John

 

[rabhishek91]

I am not quite sure what the problem to be addressed here is. Won't the driver know whether the mass in the passenger seat is human or not?

My car gives a warning when there is an object in the front passenger seat and the seat belt is not clipped (something like, "passenger air bag disabled"). It is just a panel light and is easily ignored.

If your car doesn't have that, then a logic circuit that would use the seatbelt switch and the seat switch with a similar warning that the driver could override might suffice.

John

Thanks for your reply sir. My motto is to make use of seat belt mandatory. This involves detecting whether the person is seated or not.
Say i do it using mechanical switch. So if i keep any heavy object on the seat, the switch gets activated. So it false triggers that seat belt has to be wrone(Inspite of human absence). Hope you got me .

 

[rabhishek91]

A mechanical switch is what is usually used.

I have recorded over 50 °C inside a car that was left in the sun. That was in England where the ambient temperature was less than 30 °C. I guess the seat surface temperature was more like 60 °C, as the black leather seats would have been too hot to sit on in shorts. A temperature sensor would be terrible to detect a person.

Yes sir. I think temperature sensor is not feasible.

 

[jpanhalt]

If your goal is to disable the automobile unless the seat belt is buckled, it is hard to envision a detector based on weight or temperature that would be foolproof, as already discussed in this thread.

I would also be extremely pessimistic about any system that would disable the car that the driver could not override. There are just too many hypothetical circumstances, perhaps ones in which a life hangs in the outcome, to not allow an override. At its simplest, all the driver needs to do as an override is buckle the seatbelt and put the object in the seat. Once you allow driver override, then the system is de facto not mandatory.

Nevertheless, a different approach to consider is detection of mass and organic volatiles (i.e., an electronic nose, "voc"). You could put the weight sensor in the seat as usual. The voc detector could be in the seat or in the back rest.

This first article is not related to your project specifically, but I found the history of scent detection quite interesting:

**broken link removed**

This article describes various approaches to electronic scent detection. I suggest reading about SAW and metal oxide detectors specifically. GLC and MS are quite sensitive, but would also be quite expensive.

https://www.electro-tech-online.com/custompdfs/2013/03/09_4536.pdf

Finally, here is a link to a manufacturer of metal oxide detectors. One might consider a back rest sensor that could detect CO2 (not a voc), which is not likely to released from inanimate objects. There are also numerous voc's associated with humans for which metal oxide detectors may already exist.

https://www.electro-tech-online.com/custompdfs/2013/03/APS_MOS_1109.pdf

John

Edit: Here are links to using metal oxide detectors for skatole and ammonia, respectively.

https://www.electro-tech-online.com/custompdfs/2013/03/imtc00.pdf
https://www.electro-tech-online.com/custompdfs/2013/03/detectingand_33.pdf

 

[rabhishek91]

If your goal is to disable the automobile unless the seat belt is buckled, it is hard to envision a detector based on weight or temperature that would be foolproof, as already discussed in this thread.

I would also be extremely pessimistic about any system that would disable the car that the driver could not override. There are just too many hypothetical circumstances, perhaps ones in which a life hangs in the outcome, to not allow an override. At its simplest, all the driver needs to do as an override is buckle the seatbelt and put the object in the seat. Once you allow driver override, then the system is de facto not mandatory.

Nevertheless, a different approach to consider is detection of mass and organic volatiles (i.e., an electronic nose, "voc"). You could put the weight sensor in the seat as usual. The voc detector could be in the seat or in the back rest.

This first article is not related to your project specifically, but I found the history of scent detection quite interesting:

**broken link removed**


This article describes various approaches to electronic scent detection. I suggest reading about SAW and metal oxide detectors specifically. GLC and MS are quite sensitive, but would also be quite expensive.

https://www.electro-tech-online.com/custompdfs/2013/03/09_4536.pdf

Finally, here is a link to a manufacturer of metal oxide detectors. One might consider a back rest sensor that could detect CO2 (not a voc), which is not likely to released from inanimate objects. There are also numerous voc's associated with humans for which metal oxide detectors may already exist.

https://www.electro-tech-online.com/custompdfs/2013/03/APS_MOS_1109.pdf

John

Edit: Here are links to using metal oxide detectors for skatole and ammonia, respectively.

https://www.electro-tech-online.com/custompdfs/2013/03/imtc00.pdf
https://www.electro-tech-online.com/custompdfs/2013/03/detectingand_33.pdf

Thanks a ton sir.
I'll do my homework and i'll keep you posted.
Sir the reason i came up with this idea is that here in INDIA the seat belt rules exists..
But,hardly 5% of the people follow it (i have some statistics about it)and the law here is not strict as in US(not even 20% of US). If you get caught, you may end up by paying Rs.100 (Approx 2$) fine. So people don't even bother it.
Hence i thought of doing this project.

 

[april]

[MODNOTE]Deleted Off Topic[/MODNOTE]

 

[jpanhalt]

@rabhishek91
I assume English is not your first language and complement you on your excellent understanding of it. In particular, it is clear that you understood the word, "Sir," has many uses. By far the most common use is to address a man in polite and casual speech. It is also used correctly in more formal situations, such as business correspondence and legal documents. Of course, in the military, Sir is used when referring to a person of higher military rank. It can also be used in referring to a man of lower rank and to non-military men as a show of respect. In aviation, you will almost never hear male controllers and pilots refer to each other by any word other than "Sir."

What is most surprising about the petty and offensive tirade by April (Post #18) criticizing your correct use of the word is that English is probably her first and perhaps only language. Why she picked this particular thread from among all the others on ETO in which "Sir" is used is a mystery.

I hope you have made progress in finding sensors for animal smells.

Regards,

John

 

[Reloadron]

Try a Google of Automotive Seat Sensors and Automotive Seat Sensor Circuits. There are a number of ways to go about it that are in use for sensing occupancy for seat belts and air bag deployment. The methods used vary depending on the application and include means to eliminate false triggering. They can be as simple as strings of strain gauges to more complex designs.

Ron