proximity sensor / car wing mirror / how?

Hi everyone,

I'm looking to make my own proximity wing mirror sensor project, but, can't pinpoint the type of sensors used.
For regular proximity sensors (parking sensors), the limit is 2,5 metres, max 3 metres.
(the kind used in this kit = https://www.ebay.co.uk/itm/New-Reve...655083?hash=item58cf37af2b:g:BF8AAOSwAYtWGN5U)


However, I'm looking to use sensors which sense up to 6 metres. Are there any
recommendations please?

thanks

Is this while you are on the move or to pick up if something is close to you while parking / stationary ?

You could have a look at microwave sensors if it is just to detect an object while driving.

Try googling " automotive radar "

This hit off the top of the list looks promising:

https://www.nxp.com/applications/so...bility/automotive-radar-systems:RADAR-SYSTEMS

JimB

JimB said:

Try googling " automotive radar "

This hit off the top of the list looks promising:

https://www.nxp.com/applications/so...bility/automotive-radar-systems:RADAR-SYSTEMS

JimB

Click to expand...

picbits said:

Is this while you are on the move or to pick up if something is close to you while parking / stationary ?

You could have a look at microwave sensors if it is just to detect an object while driving.

Click to expand...

This is for on the move
Looking for something low tech, i.e. Is a car there, within 6m, not necessarily NXP grade of industry leading accuracy

OEMODM said:

Hi everyone,

I'm looking to make my own proximity wing mirror sensor project, but, can't pinpoint the type of sensors used.
For regular proximity sensors (parking sensors), the limit is 2,5 metres, max 3 metres.
(the kind used in this kit = https://www.ebay.co.uk/itm/New-Reve...655083?hash=item58cf37af2b:g:BF8AAOSwAYtWGN5U)


However, I'm looking to use sensors which sense up to 6 metres. Are there any
recommendations please?

thanks

Click to expand...

Hi OE,

Your best bet would be to use an ultrasonic transmitter/receiver- the transmitter sends an ultrasonic pulse which is reflected by any object in the beam. If there is no object there is no reflected signal.

The reflected signal is detected by the ultrasonic receiver.

Electronics would then be required to calculate the time from transmission to the time of reception and from that time the distance of the reflecting object can be detected.

The simplest, lowest cost, quickest to implement, and best performing option would be to use a single board microcontroller (Raspberry Pi or Arduino) and an ultrasonic hat/shield.

The presence of an object could be indicated by illuminating a LED or a buzzer or both.

Of course, some software would be required, but nothing demanding.

spec

spec said:

Hi OE,

Your best bet would be to use an ultrasonic transmitter/receiver- the transmitter sends an ultrasonic pulse which is reflected by any object in the beam. If there is no object there is no reflected signal.

The reflected signal is detected by the ultrasonic receiver.

Electronics would then be required to calculate the time from transmission to the time of reception and from that time the distance of the reflecting object can be detected.

The simplest, lowest cost, quickest to implement, and best performing option would be to use a single board microcontroller (Raspberry Pi or Arduino) and an ultrasonic hat/shield.

The presence of an object could be indicated by illuminating a LED or a buzzer or both.

Of course, some software would be required, but nothing demanding.

spec

Click to expand...

wonderful stuff. thank you
what kind of distances, can i expect from this?
what kind of generic sensor should i go for?

spec said:

Hi OE,

Your best bet would be to use an ultrasonic transmitter/receiver- the transmitter sends an ultrasonic pulse which is reflected by any object in the beam. If there is no object there is no reflected signal.

Click to expand...

How would that work at speed with wind interference / doppler shift with fast moving vehicles etc ? You may well also have the issue of interference on rainy days.

picbits said:

How would that work at speed with wind interference / doppler shift with fast moving vehicles etc ? You may well also have the issue of interference on rainy days.

Click to expand...

good point my friend.

the sensor needs low accuracy? in the sense than it detects large objects only (i.e., a car) - not sure how that should be quantified on a sensor spec sheet
wind.......that I don't know...

Some interesting reading here : https://www.pepperl-fuchs.com/global/en/25518.htm

I was involved in a project similar to what you were after many years ago and investigated the ultrasonic side of things but there were too many variables at speed to accurately or reliably measure distance so gave up on that. You also have to bargain in other drivers ultrasonic sensors as they will also give you false positives.

Hi PB

picbits said:

How would that work at speed with wind interference / doppler shift with fast moving vehicles etc ? You may well also have the issue of interference on rainy days.

Click to expand...

picbits said:

Some interesting reading here : https://www.pepperl-fuchs.com/global/en/25518.htm

I was involved in a project similar to what you were after many years ago and investigated the ultrasonic side of things but there were too many variables at speed to accurately or reliably measure distance so gave up on that. You also have to bargain in other drivers ultrasonic sensors as they will also give you false positives.

Click to expand...

I did consider some of the points you have made but thought they could possibly be overcome, bearing in mind that this would not be a precision device.

I suspect that interference from other ultrasonic transmitters could be filtered by using pulse coding and correlation.

But if you have tried an ultrasonic approach and found it unsuccessful that is good enough for me as I have not tried it.

Thanks for the information.

spec

so ultrasonic is a no go then?

OEMODM said:

so ultrasonic is a no go then?

Click to expand...

I would now say ultrasonics are no go.

Maybe radio frequency (RF) transmitters and receivers would be more appropriate.

I'm no RF man but the problem with RF may be getting a narrow enough beam with a sufficiently small antenna.

Perhaps a microwave frequency, chosen for immunity to rain drops, might be the best way (a high frequency would allow a small printed circuit antenna to produce a narrow beam).

Is this a safety-critical system?

spec

Kids today. The side mirror wasn't even a requirement on cars till the 1960's then only the drivers side. Now days it's too much trouble to even turn their head too look at one. Guess it takes time away from texting.

While I'm at it, you kids get off my lawn.

spec said:

Hi OE,

Your best bet would be to use an ultrasonic transmitter/receiver- the transmitter sends an ultrasonic pulse which is reflected by any object in the beam. If there is no object there is no reflected signal.

The reflected signal is detected by the ultrasonic receiver.

Electronics would then be required to calculate the time from transmission to the time of reception and from that time the distance of the reflecting object can be detected.

The simplest, lowest cost, quickest to implement, and best performing option would be to use a single board microcontroller (Raspberry Pi or Arduino) and an ultrasonic hat/shield.

The presence of an object could be indicated by illuminating a LED or a buzzer or both.

Of course, some software would be required, but nothing demanding.

spec

Click to expand...

As a matter of fact I have been busy with just this typeof project the last few days. See here for one of the best sensors available: https://www.aliexpress.com/item/US-.../2038168189.html?spm=2114.13010308.0.0.O9ulqD
Temperature compensated, range of about 5m,..

Problem with longer ranges is that the sound signal gets "distracted" by all sorts of echos; the typical beamwidth is between 15 and 30 degrees, but the further you try to reach the more noise you get in the response signal.

If you are interested I can send you a summary for a program that drives a sonar and reads its responses but also removes spurious signals and does some signalprocessing to get a decent measurement.

picbits said:

Some interesting reading here : https://www.pepperl-fuchs.com/global/en/25518.htm

I was involved in a project similar to what you were after many years ago and investigated the ultrasonic side of things but there were too many variables at speed to accurately or reliably measure distance so gave up on that. You also have to bargain in other drivers ultrasonic sensors as they will also give you false positives.

Click to expand...

Other sensors only interfere if the channel on which the sensor transmits and receives is not selective enough. I know of set ups where 5 to 6 sensors spaced 20 cm apart simultaneaously measure distances without interference.

Read the Arduino and robotics forums on this subject.

Rain, wind etc.. do not interfere with ultrasound signals on short distances.

earckens said:

As a matter of fact I have been busy with just this typeof project the last few days. See here for one of the best sensors available: https://www.aliexpress.com/item/US-.../2038168189.html?spm=2114.13010308.0.0.O9ulqD
Temperature compensated, range of about 5m,..

Problem with longer ranges is that the sound signal gets "distracted" by all sorts of echos; the typical beamwidth is between 15 and 30 degrees, but the further you try to reach the more noise you get in the response signal.

If you are interested I can send you a summary for a program that drives a sonar and reads its responses but also removes spurious signals and does some signalprocessing to get a decent measurement.

Click to expand...

I must admit that I am a bit out of my depth on the practical side of ultrasonics, So any information that you have would be interesting.

spec

If the OP could provide some concrete information ie type of vehicle and exact purpose of sensors ("..on the move..": at what speeds, what purpose,..?) I would be able to provide more relevant information.

U.S. sensors can measure distances very accurately but there ate a number of constraints for use in open air. For example, no obstruction is allowed between sensor and object, so how would you accomplish this on the front side of a car?
I know of a lot of applications where these sensors are used in stationary and/or relatively slow moving vehicles but some more information from OP would be helpful.
Given the fulfilment of some basic constraints I would opt hands down for this kind of sensor for distance measurement (if you choose the right one).

Programming too is "very" easy, even including signal processing to eliminate erroneous readings (if you know C++ a bit).

picbits said:

How would that work at speed with wind interference / doppler shift with fast moving vehicles etc ? You may well also have the issue of interference on rainy days.

Click to expand...

I doubt it's a problem. Ultrasound works at speeds of around 760 mph. The issues you're bringing up might effect only a few feet per second -- the percentage difference renders it negligible.
As far as rain goes, the beam width of the ultrasonic signal and its reflection is way wider than a drop of rain. Inherent filtering takes care of any other noise.

earckens said:

See here for one of the best sensors available: https://www.aliexpress.com/item/US-.../2038168189.html?spm=2114.13010308.0.0.O9ulqD
Temperature compensated, range of about 5m,..

Problem with longer ranges is that the sound signal gets "distracted" by all sorts of echos; the typical beamwidth is between 15 and 30 degrees, but the further you try to reach the more noise you get in the response signal.

If you are interested I can send you a summary for a program that drives a sonar and reads its responses but also removes spurious signals and does some signalprocessing to get a decent measurement.

Click to expand...

1 - Thank you. These sensors are limited to about 4.5m. The sensor itself can't be covered, so, it will fail when exposed to water (splashes, rain, etc). If I wanted longer range, presumably I'd replace the actual sensors with longer range, I've had a quick look on digikey.
2 - Yes please, send me the code. I'll be using C, with a low cost MCU

>>>

earckens said:

If the OP could provide some concrete information ie type of vehicle and exact purpose of sensors ("..on the move..": at what speeds, what purpose,..?) I would be able to provide more relevant information.

U.S. sensors can measure distances very accurately but there ate a number of constraints for use in open air. For example, no obstruction is allowed between sensor and object, so how would you accomplish this on the front side of a car?
I know of a lot of applications where these sensors are used in stationary and/or relatively slow moving vehicles but some more information from OP would be helpful.
Given the fulfilment of some basic constraints I would opt hands down for this kind of sensor for distance measurement (if you choose the right one).

Programming too is "very" easy, even including signal processing to eliminate erroneous readings (if you know C++ a bit).

Click to expand...

Thank you again. I'd like to make 2 applications.

First application = Sensor stuck on rear bumper, facing backwards. If car is tailgating too close (less than 4 metres), a warning light comes on.
To be used at all speeds, all conditions

Second application = Basic after market wing mirror blind spot detection. Self adhesive fix on to the wing mirror. Facing 45deg away from vehicle. If car is in blind spot, yes or no.
To be used at all speeds. All conditions (rain, sun, etc).
This is not to be relied on, but, merely a back up warning helper. The kit might include 2 or 3 of these as a safety factor.

>>

hyedenny said:

I doubt it's a problem. Ultrasound works at speeds of around 760 mph. The issues you're bringing up might effect only a few feet per second -- the percentage difference renders it negligible.
As far as rain goes, the beam width of the ultrasonic signal and its reflection is way wider than a drop of rain. Inherent filtering takes care of any other noise.

Click to expand...

Thank you. This makes alot of sense. This would need double checking on the actual sensor datasheet (as opposed to the completed PCBA).
Also, I'm not looking for absolute precision, just, "is there a car there", yes or no, for both applications.

>>

Maybe a laser range finder type sensor? https://www.google.com/search?q=laser+range+finder+sensors&ie=utf-8&oe=utf-8 They do use them in robotics too.