This project allows to measure the speed of cars / bycicles using Radar technology, reporting the results both locally on a LCD and as well remotely on Android and on iPhone using Bluetooth LE4.0

This project address the need of measuring the speed of objects moving at variable distance.

Such things exist already on the market for below 150€ and are used not only to detect fast cars, but also for measuring the speed of e.g. golf balls or in other sports.

 

In my case, I wanted to not only count how many cars / bycicles passed a specific place, but also their speed, and then use this to correlate with the quality of air, and I wanted to do it "from scratch" for way below 50 €.

 

Several different methods exists to measure the speed of objects, and several tutorials exists on how to achieve that with Do It Yourself approaches.

 

Ultrasonic sensors 

This method is very well described, and easy to build, including a nice "radar" visualisation. (see link)

Unfortunately, however this is limited to very small range.

 

Laser

This method enables precise distance measurements however is very much dependent on the weather conditions, illumination, reflection and absorption of the materials of the target.

 

LIDAR

That is the latest approach, however unfortunately the LIDAR sensors are available starting at 40 €.

 

24GHz Radar sensors

InnoSent has been founded in 1999, in Germany, and is a market leader in the field of Radar technology. The IPM165 is their  24GHz Doppler module with an asymmetrical wide beam for detection of moving objects.  This component is used widely within a variety of larger systems.

 

Comparison with other similar project

When I started this project I wanted at first simply to reproduce what Tony Butterfield  in its DurableScope blog reported in 2014-2016 (see link). That approach used the stereo sensor RSM2650 (that is the commercialised version of IPS265) and uses for the digitalisation an external USB Sound card. Unfortunately I had large problems in making the raspberry using that data source and therefore after months of trial I stopped that approach and followed a different one, that I describe here.

 

This approach

The model is very easy to do, simply connect appropriately the parts and it will work.


The main advantage of this approach compared to the one described by Tony is the simplicty: no USB card is required, the digitalisation is performed within the Arduino that also takes care of providing the various components with their required logic level:

  • HM-10 (Bluetooth) 3.3V
  • LCM1602 (I2C LCD driver) 5V
  • IPM165 (Radar doppler sensor) 5V
  • Arduino NANO 9V

To avoid the initial problems I had with detecting signal with the IPS265, I decided to purchase a preamplified IPM165, and then to connect it directly to Arduino due to its capability of handling analog signals much better than Raspberry. When it came with visualisation, using an I2C-extended LCD greatly simplified the wiring. Networking to mobile devices using Bluetooth seems a trivial thing to do, however I had to learn "the hard way" that HC-05 is not supported by iPhone and therefore using HM-10 was the final result.

 

Test field

I tried it in my town and it works nicely, however I have not yet done a proper calibration and test. 

 

Further developments

This is only version 1.0: I will soon move to using IPS265 and amplifying it, and then in couple of months will use also try a Lidar sensor. 
In the final implementation I will also deliver a 3D model of the final case.

 

Stay tuned !