Jump to content
Sign in to follow this  
jpnorair

Parking Sensors

Recommended Posts

I'd like to set up my own parking sensor network in the area around my apartment.  I have the wireless backhaul part of it sorted, but I need to figure out what to do about the sensors. 

 

I've encountered some parking sensors, but I haven't ever stolen one off the street and taken it apart.  The concept of operation, which is the same I intend to use, is to place one sensor for each parking space (yes, I'm going to deploy over a hundred of these things).  Optimally, I can glue each unit inconspicuously to the side of the curb.

 

Does anyone have any idea what type of sensors I should be using to identify cars entering and exiting a space?  I will share the sensor design and firmware, assuming we can identify how to do it.  :)

Share this post


Link to post
Share on other sites

Why not an ultrasonic sensor mounted under the eave, pointed down at the pavement?

In normal operation, it'll pick up the distance from the sensor to the pavement- a fixed distance.

A vehicle driving in will cause a reflection from the bumper or hood and thus a reduction in distance, no?

And then when the vehicle leaves the space, the distance detected by the sensor will increase?

Share this post


Link to post
Share on other sites

I would think Inductance loop sensor would be most reliable as traffic light controls uses them. And then you have to have one under each parking spot. I don't think it's practice unless you own and operates a parking lot.

 

If I have access to a vintage point over the space, I would use a camera module (eg. OV7670 etc) to monitor a wide area, say 10-20 parking spots, and monitor for change.

 

For maximum reliability, one can paint each parking spot a yellow square in the middle and have software pick-out those as "empty" spots. Of course yellow cars we will treat them as transparent.

 

Camera + mcu processing power is more cost effective (and gaining reliability) in many application now. I can remember seeing the "older" high-end vehicles has ultrasonic transducers on their bumpers for distance sensing. And more recently, cameras are used instead, this includes rear parking warnings, lane departure warning, and forward collision avoidance.

Share this post


Link to post
Share on other sites

@@simpleavr Good thoughts, but cameras and heavy processing are not feasible for this project.  I'm trying to monitor street-side parking spots, pirate-style (if you must know, around here).  That is, it must be inconspicuous and battery powered because I don't want SFMTA getting wise to it (or anyone else).  The RF + solar + battery device I have all sorted out. What I need to do is add a sensor to that device, plus some firmware for processing this sensor data, for the purpose of figuring out if there is a car nearby.

 

@@abecedarian I'll look into ultrasonic.  That's an interesting possibility, because I could perhaps direct the ultrasonic transceiver in such a direction that it reflects off the side of the car, but when no car is there it gets no range information (sound goes off into space).

Share this post


Link to post
Share on other sites

I'd worry about street sweepers brushing away your sensors.  Those brushes are often stainless steel or really stiff plastic.

 

Really cool idea, though.

Share this post


Link to post
Share on other sites

I'd worry about street sweepers brushing away your sensors.  Those brushes are often stainless steel or really stiff plastic.

 

Really cool idea, though.

I thought of that too.  It will be easy to test, but the main device is relatively small, low-profile, and tough, so I am optimistic.

Share this post


Link to post
Share on other sites

Cool project, I second the ultrasonics too, sounds pretty robust.

 

Solar? Sounds complicated seems like if you can keep your sensor power down, you could last a year or two on a pair of AA's.

Share this post


Link to post
Share on other sites

Something like the reflectors seen on the road, but maybe made of aluminum? The shape, sloping down at the sides would deflect the street sweeper brushes around the unit. I could see a base adhered to the road bed or side of the curb, and the other section housing the detector which would attach to the base with a 1/8 turn twist- think like how smoke detectors mount to a baseplate on the ceiling.

 

The ultrasonic part could be installed so as to point at a 45* upward angle to the street and chirp every 5 minutes or so, and any reading indicating something greater than 3 feet away is ignored so that passing vehicles aren't counted as parked cars?

Share this post


Link to post
Share on other sites

Cool project, I second the ultrasonics too, sounds pretty robust.

 

Solar? Sounds complicated seems like if you can keep your sensor power down, you could last a year or two on a pair of AA's.

The solar part is already engineered.  It uses the TI BQ25504 manager IC, and possibly in the future I'll update it to the newer (but still not readily available) BQ25570.  The solar cell is a small monocrystalline device designed to work best in outdoor light.  The battery is a small Li-poly pack.

 

One other consideration is to use IR distancing.  I don't need especially accurate distancing, mostly just 0, car not parked, and 1, car parked.  IR may well allow a more rugged mechanical design.  My inclination with the enclosure is to keep it simple and sealed.  Adding vents (maybe needed for ultrasonic?) and battery doors is something I would rather avoid.

Share this post


Link to post
Share on other sites

Possibly IR, would worry a little about dirt but I guess that the same is true for ultra sonic. I think that it would be possible to seal ultrasonic too similar to how a fish sonar is constructed.

 

Regardless, probably best to test out a few things to see what works best.

Share this post


Link to post
Share on other sites

@@Lgbeno Good point on the dirt.  I'll dig around a bit to see what is possible for enclosing ultrasonic transceivers.  Looking at the energy requirements of ultrasound (not too bad), I would estimate that I could probably implement range checks each 10 seconds during peak hours (or if there is surplus solar energy), and once per minute or two during off hours.  I really do like that the ultrasonic transducer has much lower peak current than does an IR LED, because my battery doesn't have high current sourcing ability.  BTW, in my universe of low-power electronics, 10mA counts as "high current."  :)

Share this post


Link to post
Share on other sites

What is the distance from car to electronics?

 

You probably already know that an inductance loop for traffic sensing is huge - 6 foot diameter by 4 turns.

 

Here's my brain storming:

1. IR: sense the heat produced

2. Inductive: sense the metal

3. Pressure: sense the weight

4. Ultrasonic: sense the audio reflections

5. Light: sense the blocking of light when car parked ontop of device.

6. CO2: sense the exhaust

7. Sonic: sense the noise (characteristic frequency of an engine at idle)

8. Video: sense the presence or absence of a vehicle in a video frame (common method).

9. Lasers: sense the change in reflection

 

On a non-permanent, sketchy installation, you are limited in your options.

 

It looks like you live beside a popular park. Could you take advantage of your building to mount any sensors that could overlook your desired parking area?

 

On a side note, I am using the same bq part in an industrial agricultural design. Have you considered using LiFePO4 batteries instead of LiPo's? 

Share this post


Link to post
Share on other sites

I did a lot of work on battery-operated parking sensors for a startup recently.  We used magnetic sensors that are sensitive enough to detect distortions in the earth's magnetic field.  The iron an an automobile distorts the field and either increases or decreases the detected field which is read from the sensor by ADC.  They're quite low power to begin with, plus you can switch them on with a transistor for low duty-cycle sampling, leading to very long operation from battery.

 

http://www.magneticsensors.com/vehicle-detection-solutions.php

 

In our project we used the HMC1021 in our case.  I don't think I'm able to share code, but they were very simple to operate, and the datasheet has nice example circuits.

Share this post


Link to post
Share on other sites

It looks like you live beside a popular park. Could you take advantage of your building to mount any sensors that could overlook your desired parking area?

 

On a side note, I am using the same bq part in an industrial agricultural design. Have you considered using LiFePO4 batteries instead of LiPo's? 

This project is more about the means than the ends.  In other words, it is important to do it with a small tag and limited infrastructure.  I suppose I could use any compliant battery.  The current device uses a small LiPo.  I would be happy to use LiFePO4, provided that I can find one in the right form factor.  I also like that it seems to have a lower voltage.

 

 

 

 

In our project we used the HMC1021 in our case.  I don't think I'm able to share code, but they were very simple to operate, and the datasheet has nice example circuits.

Thanks.  I'll check that out.  As mentioned, one major goal is to do build a parking sensor that fits entirely in the curb.  The solar power budget is actually much higher than what a typical primary-powered battery sensor has, so I do have a few options.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  

×
×
  • Create New...