#198

Laser range sensors can measure long distances with high precision. Today we try to build not only such a range sensor. We want to use it also as laser speed gun. Will it better than our radar sensor?
In video #181 we tried to measure speed using a cheap radar sensor. This project is still in the lab, and I expect an update in the next month or so. In the meantime, I started experimenting with another technology we already encountered in video #119: Time of Flight sensors used in LIDAR systems. This was one of my longtime wishes, and it became reachable when I saw these new rangefinder modules pop up on AliExpress. They have no display but a digital interface. Exactly what we need for our hack.
Links:
Range Finder: bit.ly/2Fnpwb9
Golf Range Finder: HTML Code:
s.click.aliexpr...
Supporting Material and Blog Page: www.sensorsiot.org
Github: www.github.com...
My Patreon Page: / andreasspiess
My Bitcoin address: 19FSmqbBzb5zsYB1d8Bq4KbxVmezToDNTV
If you want to support the channel, please use the links below to start your shopping. No additional charges for you, but I get a commission (of your purchases the next 24 hours) to buy new stuff for the channel
For Banggood bit.ly/2jAQEf4
For AliExpress: bit.ly/2B0yTLL
For ebay.com: ebay.to/2DuYXBp
www.facebook.c...
/ spiessa
www.instructab...
Please do not try to Email me or invite me on LinkedIn. These communication channels are reserved for my primary job
Equipment in my lab: www.sensorsiot....

[@northshorepx]

I do believe that Andreas is building skynet. Seriously though. This is not a 100% failure. You have given many people sound knowledge on how to conduct experiments soundly and scientifically. Using this tutorial people can start to think about how to improve their own skills and hopefully not be taken in by lots of false advertising.

[@AndreasSpiess]

You are right. Failure often has more learning in it than success...

[@Peter_S_]

Successful experiments often involve production of negative results.

[@korishan]

There is no such thing as failure in these types of things. You just found another way that it doesn't work. 😎

[@waltsteinchen]

Failures do not really exist. They are always a part of learnings .. and in case of youtube it's very helpful for other because the know what already failed .. or maybe sometimes they have an idea what could resolve the problem. And Andreas always catches up those hints .. that's why I really like his videos.

[@AndreasSpiess]

You are right. I wanted also to warn others not spending the money...

[@TonySwitzerland]

Hochinteressante Beitrag, fesselnd präsentiert. Klartext! Das unerwartete Schlussresultat tut der Sache keinen Abbruch. Nein, es zeigt einfach die momentanen Grenzen auf und wie gut man den chinesischen Spezifikationen glauben darf. (Hab da einige Dispute hinter mir)

[@AndreasSpiess]

Nach einigen tausend Aliexpress Bestellungen muss ich trotzdem sagen: Für mich hat sich das alleweil gelohnt. Nur schon wegen der Auswahl an günstigen Teilen. Und die (gut dokumentierten) Disputes wurden auch fast alle zurück bezahlt... Dieser Kanal würde nicht bestehen ohne Aliexpress und sein weltweites Distributionsnetz. Das kann nicht einmal Amazon.

[@douro20]

This is a very difficult problem and one which has only been solved by a handful of companies. The first was Kustom Electronics, a Kansas company which originally built electronic musical instruments and amplifiers. Using experience they gathered from development of the US Space Surveillance Network, they set out to develop the world's first handheld laser speed detection system back in the mid-1980s. Their equipment is considered the gold standard for LIDAR speed detection.

[@AndreasSpiess]

Thank you for the information. In the newest video, I tested a better device ;-)

[@Wookey.]

I'm a cave surveyor and have been using these lidar devices since they were really expensive and the size of a shoebox back in the early 1990s. The way they do the impossible time-of-flight measurement is by modulating the laser signal, and then using the pattern of the mixed direct and reflected signals (much like you did at the radar frequrencies just with the clever signal processing). The tech was invented by Leica originally. So essentially you actually use the modulation frequency for analysis, not the raw light frequency, which as you say is impossible to measure directly. Early models had a physical optical splitter on a solenoid, but that became solid state as they got smaller and cheaper and lower power.

[@AndreasSpiess]

You are right. I later found out that this is one of the concepts used by the devices. The newest ones seem to be able to really measure TOF without the trick of phases.

[@Orbis92]

I find the application notes of the ToF chip GP-22 very helpful. I learned a lot about laser ranging techniques from them. For example their GP-22 uses phase shift for mm-accuracy and for km-distances they use normal time-of-flight. Internally the use some kind of GHZ ring oscillator to count pulses between start and stop signals :)

[@AndreasSpiess]

Thank you for this excellent link. It helped me a lot. After more investigation, I found that my sensor works by sending signals of different frequencies and measure the phase shift. I assume the range finder in the mail will use this GP-22 chip. We will see...

[@bardenegri21]

Picking up a comercial unit and sniffing the signal between the sensor and the micro indeed sounds like the way to go. Very interesting video, thank you.

[@AndreasSpiess]

I still hope one of my viewers knows more about the principle. It should be public knowledge, I think. With all this autonomous driving...

[@CTCTraining1]

I’m pleased you are challenging the seller for a refund ... so many channels just stop at the point they have a video complaining of the quality. I hope your thorough and diligent engineering approach will be rewarded with great results (and a full refund!).

[@AndreasSpiess]

I always apply for refunds, because I think, without feedback, this system does not work. But it is more work for me. BTW: I got my money back in nearly 100% of the cases. I also have to say that I have a low percentage of disputes.

[@DougHanchard]

Testing is always the preferred method of determining accuracy of a products claim. Really enjoyed this video as Andreas points out important issues. It's how the scientific community has create physical laws we all accept. There might be some reasons why the "laser" is not working properly, including the wavelength configuration and the receive sensor. In fact, it may not even be possible to operate as desired if both are not calibrated correctly. Light is a very finicky "thing". We know this from decades of fibre optic wavelength testing. The power, precision, and tuning demands are significant. And this is in a closed and controlled environment! In an open environment, the obstacles and dilution (scattering) effects that can obstruct the wavelength / spectrum require a really good receiver sensor or failure is likely. Light can be your friend or your worst enemy. The only observation I have is processing cpu power on the board and the data transmission rate between the serial port and the laser detection settings. The number of queries might be overwhelming the serial port and a reason why latency is occurring and increasing with "time". You might want to inject a interval and observe the results. #impedance spectroscopy

[@AndreasSpiess]

It seems you have a lot of experience in this field. I do not know a lot. But I expect that the supplier knows and ships me a device which can fulfill the specs he promises. As I mentioned I have another rangefinder on order and we will see if this one is different. For me, it is still hard to believe that such devices can "see" a small point over 100 meters and more...

[@DougHanchard]

I find it unlikely you will find the performance you are looking for at the price point and range desired. There are many technical requirements for a laser to perform at 50 to 100 meters using a power voltage range of less than 5 Volts (even if it is pulse boosted to 10 Volts) that registers consistent measurements that must be processed using ESP32 boards. And then there are the laws of physics that must be accounted for. Using a laser / LIDAR for speed detection requires precision components. Consumer grade speed guns do not use laser frequencies and are incredibly accurate in very small form factors. Bushnell makes one that's smaller than a 9mm Beretta and runs on two AAA batteries for less than $85 USD. It has a range of over 1000 meters. As you have discovered, the geometry and wavelength filtering (signal control; reflection, refraction, diffraction, absorption and scattering) issues are staggeringly complex that demand intensive processing power at VERY fast speeds - assuming the quality of the components are superior. The send / receive components are not cheap and require multiple processing steps before the data can be used with any degree of accuracy. A 600 DMIPS (Dhrystone Millions of instructions per second) logic controller that does not have floating point calculation capability running at 240 MHz is probably going to be pushed to its limits when used on a multi-purpose ESP32 board. It also requires multiple computational threads for signal input and output data if for any target data that is moving. If the device is not held perfectly still, it too will affect performance and your results. You are also faced with the laws of physics and the reaction of soft materials (clothes, human tissue, etc.) that tend to absorb light while dense hard materials (steel, wood, etc) will have better reflection properties.

[@DougHanchard]

Andreas Spiess You may want to take a look at this unit. kzbin.info/www/bejne/qIGzdaCch9ChrKc

[@PhG1961]

Of course it was at least entertaining ! Your remark near the end of the video about the capacity of batteries made in China... is so true.

[@AndreasSpiess]

And at least I learned a lot about measuring distance using light...

[@BrianBoniMakes]

I enjoyed watching this and was hoping for success as I've suffered a similar problem with radar speed guns but haven't tried laser.

[@AndreasSpiess]

Mee too was hoping for success... But I will give it a second try. Meanwhile I have some more „weapons“ in the lab

[@zvikamikaze]

Excellent post, as always, thank you! And indeed the method is so much more important than the results. Especially, I really loved the TX trick to find logical level voltage. One comment: perhaps one more thing that can be done to augment the device is replace the weak laser diode with a stronger one, operating at the same wavelength. If possible, I would disconnect the existing laser and hook up a beefy mosfet feeding one of the stronger laser emitters. I wonder what does the sensing of the reflected light, though, whether it is the emitting diode itself or a dedicated sensor

[@AndreasSpiess]

I thought about that. But it seems to be a coupled system where not only the reflected light is used but also the emitted one. And the diode is not only switched on and off. A quite complex voltage curve is driving it. I also want to see if the strength of the diode makes the difference between this one and the one on order.

[@zvikamikaze]

Well, that's interesting. Good to know we had the same line of thinking. I wonder about the theory behind that little fellow...

[@justinmallaiz4549]

Glad to see your back to sensors 😊 I remember you mentioning you would like to tackle detecting bicycles. I hope your still motivated to tackle this difficult project. (I believe you’ll be forced to run a neural net using a cheap camera sensor. ) my fingers are crossed your vids go in this direction 👍

[@AndreasSpiess]

My trainee is still working on it and maybe, we will be able to show it in the near future.

[@lmamakos]

Possibly a filter on the detector element on the receiver tuned for the wavelength of the laser to avoid swamping the detector with more ambient light than necessary? Unless the detector already has some sort of filter already in use..

[@AndreasSpiess]

The detector seems to have a filter on it. But ambient light also has parts in the IR wavelength...

[@andrewhussey2002]

Entertaining and educational video as usual; thanks. Could I perhaps suggest a video on your advice for purchasing from the Chinese websites for components? More pointers on what to look out for in the specifications and how to chose for those of us just starting out maybe?

[@AndreasSpiess]

I do not have tricks. I use Bangggod or Aliexpress, and if something does not work, I create a dispute. Usually, I get my money back or a new product.

[@pksdematerializedbrain7600]

Good one .. would have been great if this project was a success. Great DIY project to try, rangefinders are fascinating :)

[@AndreasSpiess]

Thank you!

[@curranhouse]

Andreas we need your Swiss accent to tell us how to jam a laser speed trap 👍🏻😂😂

[@AndreasSpiess]

We do not drive too fast here :-))

[@curranhouse]

@@AndreasSpiess of course :)

[@Ed19601]

Going to the gym for broader shoulders is a possibility. Eating more to get a bigger belly is more efficient though :-)

[@AndreasSpiess]

You are right. But whether Mrs. S. would see it the same way?

[@pierrec1590]

Makes me wonder if you couldn't put the light sensor behind the binoculars to get better signal and reduce interference from ambient light. I do agree with the other viewers who think it was NOT a total failure.

[@AndreasSpiess]

I will compare this one with the one I will get. The new one claims a 6x magnification.

[@BerndFelsche]

Laser speed guns use infrared and that can be filtered even during daylight... the "spot" is not visible to the naked eye. I had a chance tolay with a proper unit many years ago and the number plates and headlights were the most reliable aiming point. The unit sent out a train of about 16 pulses in a third of a second and did a simple average to determine speed. I deliberately swept the point of aim between walls on the opposite side of the road and got the building to move at several kilometers per hour. 😎 Base technology and firmware has improved over the years so such shenanigans are much harder now.

[@AndreasSpiess]

Thanks for the info. I should get another one for experimenting and for sure will give an update.

[@superdau]

My understanding of how these laser "tape" measures work is that they vary the laser frequency and measure the intereference with the reflected signal. During this frequency sweep (easily done for semiconductor laser by varying the current) you will have multiple destructive interferencies (which are dead easy to detect; pure analog circuitry is enough). The "distance" (in wavlength) between those destructive interferences is dependent on the distance to the measured object. Since the modulation sweep speed of the laser can be whatever you design it to be, you can take whatever speed you need to measure the time between those destructive interferences. This would also explain why the measurement drifts when the laser is on for a longer time. The laser diode will heat up (which changes the wavelength), but the circuit is still assuming a base wavelength of say 650nm.

[@AndreasSpiess]

You are partially right. I did some more investigations and found the laser is modulated with signals of different frequencies. Like that they can measure the phase difference and calculate the distance. Because light has a very short wavelength it would not help shifting frequency there (distance between maximums would only be in the nanometers). If they would use 100 MHz the "wavelength" is 3 meters and they could distinguish all distances between 0 to 3 meters. We measured some frequencies around 200 MHz.

[@PsiQ]

the laser speed guns also need to be aimed at the number plates afaik, which is not that easy. Also problems with de-flections from windows + curved surfaces. The cloth reflector material probably is optimized for a different light frequency, or might even only reflect a specific nm range in the visible spectrum.

[@AndreasSpiess]

You are probably right concerning the reflectors. I did not think about that.

[@MeNeXzPnz]

Im wondering if you could build something like this using the interferance of the laser with its reflexion for different frequencys of the laser pulses

[@AndreasSpiess]

Maybe you watch my video about LIDARS. Some Rangefinders use interference signals (phase)

[@AirCro]

Dear Andreas, first i want to say love your work and keep it up especially LORA videos. Second since your video 181 i also had the idea of a laser speed detector however my approach is defrent. I have a proplem of fast drivers going by my house i wanted a stationary speed trap. The idea is to put two laser points on one side and laser reader with esp on the other side of the street. Knowing that i will place the sensors and laser pointers 10cm apart i can calculate the time from brack of one laser been to the time of brake of the second and calculate time to get the speed. I had a small sample built but i would like to add a camara to catch the plates as well. However i dont have the time, knowhow and the resorces to make it. Maybe you can add this type of speed senser to your video next time.

[@AndreasSpiess]

Yours is a simple principle but needs more installation. It is very simple. Use two sensors, start the timer with one sensor and stop it with the second one. Then you can calculate the speed...

[@christianbartsch]

Have you found a better laser component to solve the problem as the original HiReed component was by far too weak for the intended purpose?

[@AndreasSpiess]

I abandoned the idea after that experience :-(

[@guidowager]

well done! Very innovative! Thanks for all the effort!

[@AndreasSpiess]

You are welcome!

[@steadfastnz]

Perhaps you could still build a speed sensor, just using 2 of these pointing across the path, a known distance apart and communicating to a node that would do the calculations? You could even measure the speed of multiple vehicles if they are travelling different lanes on a road, just by measuring the distance to the vehicle from the sensor.

[@AndreasSpiess]

That would be a different principle. Probably not easy because there is not a constant beam. there is only a measurement every 0.3 seconds. So the resolution would be rather small and the distances between the two devices would have to be quite long.

[@JanCiger]

Andreas, these ToF sensors don't really measure time of flight but instead send out a modulated signal and measure a phase difference between the original signal and the reflected one. The phase difference is directly proportional to the distance. That is much easier than to measure time. So time of flight is a bit of a misnomer.

[@AndreasSpiess]

You are right with the method (I did some more investigation after the video). But they measure the "time of flight" with this principle. You can check it by including another material like water into the beam. It changes the "distance" because the speed of light is different.

[@JasonWho]

Still waiting for the final video on the Doppler Radar speed detector, please circle back to it! Thank you!

[@AndreasSpiess]

This project is sleeping. But it might "wake up" when the winter is over as I got a few new modules...

[@MrMaxeemum]

I have always wondered about a speed gun which worked from the side. Using a standard camera select 2x spots (left and right) on the image and detect when they change with a time stamp. Then drive past at a known speed to calibrate it (could also measure the distance between the 2x spots. I've never got around to doing this though.

[@AndreasSpiess]

Your method works perfectly. However, it needs some installation (4 poles) and its limitation is if two cars pass in two different directions at the same time.

[@VerstehenSieMathis]

Some of these meters use laser-triangulation: The sent laser beam (known angle), the beam coming back, and the (known) distance laser->optics form a triangle. Behind the optics is a sensor which can tell where the beam hits it, which is depending on the angle of the beam coming into the optics. So we know two angles and one distance; rest of the triangle can be calculated.

[@AndreasSpiess]

Triangulation is usually used for short distances. I cannot imagine that the base of 2cm can be used for a 50 m measurement. But I might be wrong, of course.

[@VerstehenSieMathis]

Yes, 100m is maybe a bit far, but it could explain your problems with heat (warping) and the different surfaces. And if the laser spot is small and pulsed (for ambient light cancelling), it wouldn't appear bright to the human eye. The mirror might be there because the sensor is mounted flat on the PCB... so I thought it kinda fits

[@VerstehenSieMathis]

Oh, also: LCD, not OLED? :-)

[@joachimfuchs3831]

Laser range sensors usually use triangulation, not time measurement (de.wikipedia.org/wiki/Elektrooptische_Entfernungsmessung#Lasertriangulation). The reflected laser point is projected on a CCD-sensor. I gues the arrangements of laser, lens and CCD and so the reliability of the measurement depend on the temperature. This would explain that reliability becomes worse after a longer time of continous operation.

[@AndreasSpiess]

Thank you for the link. Reading this interesting Article I would say, that it uses TOF plus modulation of the laser. I looked at the voltage across the diode we saw a high frequency signal which would explain the modulation part.

[@joachimfuchs3831]

the spec on aliexpress says 100m, not 1000m. But you can't trust this. And the range like Andreas found out was only a few meters. For longer distances you are right. Triangulation won't work here if emitter and tranceiver are that close to each other.

[@joachimfuchs3831]

OK, may be you are right. But the modulation may also be used to distinguish the reflected laser light from incoming ambient light. What is the frequency you measured?

[@AndreasSpiess]

There were a few signals overlapped. But with what I learned now I have to check it again and measure the current, not only the voltage.

[@bostonmacosx]

Any update to the Radar as a speed sensor. Would be interested in the integration with ESP32? how do you connect the two?

[@AndreasSpiess]

We never continued the project :-(

[@bostonmacosx]

@@AndreasSpiess I'm going to with my limited knowledge try to hook up a MAX4466 or LM358 to amplify the signal and then try to FFT the in and get some speeds ;)

[@codeveloper3095]

Andreas Spiess, ein sehr interessantes Projekt wäre ein Datenübertragungskabel mit Laser, klar es gibt Internet Glasfaser, aber ein DIY Laser USB Kabel oder allgemeines Datenübertragungskabel mit Glasfaser und Laser wäre echt interessant :D

[@AndreasSpiess]

Da verstehe ich gar nix ;-). Aber ich habe ein Projekt wo ich versuche, mittels starken IR Dioden eine Datenübertragung auf eine grössere Distanz zu kriegen (durch die Luft). Aber ich habe keine Idee ob das geht. Das Material ist jedenfalls hier...

[@codeveloper3095]

Naja imprinzip funktioniert es genau so wie du es mit den IR Dioden vor hast, denn Datenuebertragung ist ja nicht anderes als High & Low Signale mit gewissen abstaenden zu schicken. Dies funktioniert ja eben auch mit einer lichtquelle(laser) und einem empfaenger der die signale des lichts empfaengt. aber du hast recht, die auf das Standard USB Protokol zu uebertragen ist extrem schwer bzw. komplex. Deine idee waere da fuer den Anfang natuerlich besser um zu verstehen wie datenuebertragung mit licht funktioniert und in wie weit man das ausdehnen kann mit der geschwindigkeit. Du kennst das ja beispielsweise bei der Seriellen Uebertragung mit den Baudraten. Da waere es natuerlich sehr interessant zu wissen in wie weit man daten mit licht uebertragen kann und in welcher geschwindigkeit. Die sache mit Glasfaser kann man ja da auch erstmal zur seite legen, denn das Glasfaser macht ja nichts anderes als das licht in dem Faser zu isolieren so das man damit daten uebertragen kann in einer extrem weiten entfernung ohne lichtverlust und geschwindigkeitsverlust, denn licht in nunmal schneller als elektronische impulse ueber lange kabel(abgesehen vom spannungsverlust bei sehr langen kabeln)

[@codeveloper3095]

Einfach mal einen laser nehmen und ein modul der die Laserstrahlung empfaengt und auswerten kann, das waere ja mal ein sehr guter anfang :)

[@AndreasSpiess]

Soweit ich weiss übertragen sie bis zu Gigabits/s über eine Faser in den Unterseekabeln. Soeweit werde ich sicher nicht gehen können ;-) P.S. Licht und "elektromagnetische Wellen" bewegen sich gleich schnell (Lichtgeschwindigkeit). Die hängt allerdings vom Medium ab.

[@MoritzvonSchweinitz]

If it's using the phase modulation for the distance calculation, I think that the optics should be replaceable? As long as the sensor receives some laser light with the modulation, it should work? So maybe you could use your binoculars to boost the range? I think you wouldn't even have to re-calibrate for the the added magnification.

[@AndreasSpiess]

You are right with the phase detection. It uses different frequencies (I made some more investigations). The new one on order should have some sort of magnification. We will see...

[@MoritzvonSchweinitz]

Thank you for your reply, and your great ESP32/ESP8266 videos! They are a great help!

[@thesimbon]

That is why radar speed sensors are widely used, no need to aim precisely and no reflective surface is needed. There are laser speed guns used like in Italy and there are countermeasures sold on the internet: they usually are a led strip to be mounted on the front of the car that hat blinds the laser sensor; of course they are illegal but you know how it works

[@AndreasSpiess]

Here they use both. Radar for short distances and laser for longer distances. Lasers are operated by a police officer, and Radars can be run automatically.

[@marciochao4849]

Hi Andreas, For short distances like 5 meters max. The diatance measurement is acceptable?

[@AndreasSpiess]

I do not know. You get other (specialized) devices for that purpose, I think.

[@zukjeff]

@3:10 sounded like " 19.2 kilo volts" but may have been baud ;-). Not a total failure, it was a practical learning in the field with measurable learnings. well done.

[@AndreasSpiess]

Maybe the effect o my accent ;-)

[@wktodd]

I suspect the little hand held devices use spot size (triangulation) to measure smaller distance and an time encoded pulse for larger ones. BTW I guess the red laser is only for siting i.e. it almost certainly uses a IR laser integrated with the camera for measurement . Again a guess, but one way of measuring distance is to vary the frequency of the send signal linearly and measure the beat frequency of the combined send/received signals .

[@AndreasSpiess]

I have to check again. But I think I remember it did not work anymore when I covered the LED.

[@wktodd]

Andreas Spiess you may be right. the red laser is modulated so it could be used as the signal carrier. I may have to play around with my one ( but first i have to fix a 30 year old studer tape deck)

[@AndreasSpiess]

Studer decks are always worth the effort. I used to work in the next building to their factory (also 30 years ago)

[@wktodd]

Andreas Spiess studer fixed (well faulty board found and bypassed) will have a look at the range finder signals tomorrow

[@science4allworld587]

Just wondering what happens if two bicycles with same laser sensors at different distances point at same target? With both be “confused”? Imagine when several of these sensors from several vehicles in traffic.

[@AndreasSpiess]

The lasers are not mounted on the bicycles. But you can confuse them. Because the laser points are quite small and the time of measurement is short it is not easy

[@MobiusHorizons]

If you choose `D` instead of `F` do you get a longer distance?

[@AndreasSpiess]

No, unfortunately not.

[@MobiusHorizons]

does it return a higher resolution at least? otherwise what is the point of the slow mode? I guess it might use less power.

[@AndreasSpiess]

As I said, I did not find a difference. Also not in the brightness of the dot. But I did not investigate too long. Because I anyway was not interested in the slow one.

[@Honeypot-x9s]

I’m betting it’s 2 factors at play here... one is it’s Chinese junk that is falsely speced.. Two is that it’s very unclear how this device is even working. It is possible that the laser maybe under powered (too low voltage/amperage) and also possible that the micro processor is getting too much power (he did mention it drifts a lot when continuously used which heat would explain). Bonus third possibility is that it was poorly engineered on silicon level, things are internally wired poorly or a lot of crosstalk making accurate readings very difficult.

[@atakankocyigit9984]

Hey man, How did u program this laser. Which code and connection. Could you tell me ı need that a lot. Thanks

[@AndreasSpiess]

If I did not post a link, I did not publish the code :-(

[@lgeo_al]

The value after distance that You obtained in YAT seems to be "intensity" value of ray that has been received. You can easily verifi that by measuring distance to black/dark object (ie. black sheet of paper) and to a white sheet of paper - intensity values should differ significantly, at the same range to the object.

[@AndreasSpiess]

Well possible. I have to check.

[@electronic7979]

Excellent

[@AndreasSpiess]

:-)

[@eddyfontaineyoutu100]

Hello Andreas, Keep on working on this nice project ! Are you aware that you switched YAT to 19.2 kVolt instead of 19.2 kBaud ? ( @ 3:10 min ) 😊😊 Cheers, Eddy.

[@AndreasSpiess]

That is probably where I killed the device without knowing it ;-)

[@brainfornothing]

Thanks for sharing !

[@AndreasSpiess]

:-)

[@MarcRitzMD]

I feel like it is time that you do your own take on Iron Man. You could integrate the range finder into making a projectile shooter which needs to take ballistic trajectories into consideration

[@AndreasSpiess]

Actually, the professional ones for hunters seem to have such a feature ;-) But a good idea for a video!

[@ahaveland]

A collab with Colin Furze perhaps? :-)

[@Antiath]

2 mm accuracy and 2cm minimum distance ? That's completly nuts ! I want to know how they can measure such small time scales.

[@AndreasSpiess]

I did not try the 2 cm minimum distance, but the +/- 2mm was ok (not absolutely, but relatively. I kept the device at the same spot and measured several times. As said, the values were very comparable over distances in my lab. And I think the normal rangefinders are also quite precise.

[@ooloncolluphid9975]

I am reminded of Michelson-Morley type experimental setups where they use interferometry, maybe it is the case here. So instead of measuring time difference it measures the phase shift ?

[@Antiath]

It could make sense but you have an very long arm ( the mesured distance) paired with a very short one ( in the device). Because of this difference, you risk loosing temporal coherence and this would result in a loss of contrast of the fringes.

[@chemieprof]

They probably don't measure time (time of flight) directly, but rather phase shift. When light travels, it also shifts in phase (because light is a wave (sometimes at least ^^)). You can use that phase shift for interference with another light source inside the device and thereby change the amplitude of a resulting light signal which then in turn can more easily be measured.

[@pinealservo]

The commenters mentioning phase shift are almost certainly right. Recent depth-sensitive cameras from PMD/Infineon have an IR ToF sensor per-pixel and give amazingly detailed 3-D contour data in real-time, but only out to 6m or so, and only at that distance with a slow update rate. The PMD development cameras are also not cheap!

[@ThisMicrophoneSoundsCheap]

I believe that these are laser triangulation sensors. Depending on the distance from the object, the beam gets reflected onto a different part of the matrix behind the big lens.

[@AndreasSpiess]

The sensor I used measures the "Time of Flight". But sensors for smaller distances can use the principle you describe.

[@ThisMicrophoneSoundsCheap]

I think your "long" range sensors use triangulation, not ToF, that would explain the big lens and a long photosensitive sensor array behind it. Also, it could answer the question why they don't work with reflective vest surface. It's designed to reflect light exactly into the "sender" and it's not what the triangulation sensor expects to happen with its laser. I'm very interested what will happen if you shine it at a mirror at slightly different angles than 90 degrees (please use protective glasses). Try to get its own laser to shine at different points of its big lens. Maybe cover the lens with a piece of tinfoil with a small hole.

[@AndreasSpiess]

I added a few other videos with some more "discoveries".

[@gabywid2004]

Hi, I was just about to buy this module, but after watch this video, now I doubt whether this module can meet my need. I'm starting DIY "forward collision warning" using arduino, OBDII (to take car speed data) and distance measuring sensor, problem is until now I can't find good distance measuring sensor up to 50 m.. If you find a module with that spec, please let me know, or maybe you interested to DIY that project :-) .... thank you in advance.

[@AndreasSpiess]

I abandoned this project.

[@avejst]

Thanks for sharing, a good start :-)

[@AndreasSpiess]

We will see. At least I learned a lot!

[@tmcdon4ld]

I don't think that any sensors outside of the lab, satellites, or maybe fighter jets actually use time of flight they compare the phase of a signal sent over the laser.

[@AndreasSpiess]

You might be wrong. But for the moment I cannot prove it. Stay tuned.

[@stylesoftware]

I bought the 3rd laser on the right recently, it gives 20 readings per second in continuous mode. But it has problems with lidar like applications. Namely like the other three lasers, when it encounters an error, it will slow down to 3 seconds for an error reading. Typically this kind of laser tries 'thin' mode, and if it can't determine the signal will try 'thick' mode. More annoyingly when modeling 3d objects, the laser shuts down every 5 seconds for 1 second, i guess due to heat. True 200hz - 500hz lasers do not do this, they simply return an error without the retry of different beam widths - but they suffer from much higher inaccuracy.

[@AndreasSpiess]

Fortunately, we can get now cheaper LIDARS which are made for things like 3D modeling...

[@stylesoftware]

I agree, I scored one for 50 USD as a fluke. But then the company changed to 150 USD :( The cheapest lidar @ 100-500 readings per second, at 3cm accuracy, seems to be in the 150 USD range. Do you know of a cheaper one?

[@jmr]

Very disappointing. I hope you crack this problem! This is probably why the Mythbusters always used cameras with markers painted in the background. I'm thinking about using two sensors set as far apart as possible along with timestamps and RTCs. Even then there would be the issue of incorrect numbers due to multiple vehicles. I wonder if we could resurrect the Mythbusters method using cameras an automation.

[@AndreasSpiess]

I stopped these projects a few years ago. Maybe I will start again when Radar sensors become cheaper (because they are used in cars)

[@jmr]

@@AndreasSpiess We may have other interesting ways to measure traffic when cars start talking to each other.

[@dajg1995]

"Maybe I have to go more to the gym to get broader shoulders" I love your humor! :D P.S.: As always, great video! :)

[@AndreasSpiess]

Thank you for your compliment.

[@denisdesrosiers1955]

Thank you so mutch for all your video . I have stupid question , i know how to upload arduino program in esp32 but there is a way to retrive it (download) back too arduino ide for revers engineering ?

[@AndreasSpiess]

You can download the bytes of the flash of the ESP32. However, they are only machine-readable (the compiler in the Arduino IDE translates your code before uploading). You would need a "disassembler" or similar to get your code back. This is a big effort for specialists.

[@denisdesrosiers1955]

Merci beaucoup !

[@AndreaZeta]

Andreas, what is the name or Oled size of the ESP32 + OLED board in this project? Thanks!

[@AndreasSpiess]

will see all boards with oled in one of my next videos. This one is from TTGO

[@nadiakhicha3892]

Hello, I have a question. Why didn't you share your project on github?

[@AndreasSpiess]

Usually, I share my projects. But this was too trivial, I think.

[@kirrshan7375]

Hello sir how can i measure distance using this sensor module and esp12E? Can you pls guide me on this. Thank You in advance.

[@AndreasSpiess]

Because I was disappointed. That is why I did not publish the code...

[@kirrshan7375]

@@AndreasSpiess Will be much appreciated if you could upload it. Need some code guide for my final year project. Please help me :) ..

[@PatrykDarasz]

Please try to built some nice night vision glasses :)

[@AndreasSpiess]

That is probably too complicated for me. At least the ones working without light. Wit IR diodes it seems to be easier...

[@shaunwhiteley3544]

Andreas Spiess Another vote for night vision glasses 😀. Please 🙏. Cheers

[@PatrykDarasz]

My point is not really a "professional view" of what you can see in the dark, of course some camera with IR diodes will do fine. The thing I am after are the actual glasses/ goggles. I want a "small device" that I can put on my head and see everything nicely. The problem is to make some good glasses that are *short* and allow you to view video. A advanced version of this would be the display inside not only showing you video but also range, temperature and so on like iron mans helmet. But you could give it a try anyways :)

[@JamsAlx]

What happened to the speed gun project Andreas?

[@AndreasSpiess]

It was abandoned, because I did not find the right parts (and also had different priorities)

[@JamsAlx]

That's a shame. Keep up the good work 👍🏻

[@tituszban]

Your friend looks like Gilfoil! (From Silicon Valley)

[@AndreasSpiess]

He agrees :-)

[@snitkjaer]

I would really like to know how this laser sensor measures time of flight with relatively cheap components? Can it be taken apart for a bit of reverse engineering? Or does somebody on this channel know?

[@AndreasSpiess]

Mee too. I did not find a lot of information about it. And I am not sure if taking it apart helps a lot. You can see most of the parts already open. I thought they use some sort of phase to determine the distance. I found an oscillator chip which is rated to max 25 MHz and a crystal close to the LED. So they need some "precise" timing.

[@anlumo1]

Maybe the voltage to the laser was too low for some reason, and that's why it didn't have enough brightness?

[@AndreasSpiess]

I did some more investigations and found that the laser is switched on and off with a frequency up to 200 MHz. So it would be difficult to change frequency without changing the whole device...

[@anlumo1]

A video about designing your own laser range finder would be a great addition to the channel!

[@muhannadabu_shaqra4949]

Can you please share the code even if didn’t work

[@AndreasSpiess]

I always share the code if anything is shareable. This device outputs ascii characters

[@zvpunry1971]

Where I live, you can find speed measurement devices including tripods on the roadside. I think that there must be somebody who has too many of them and just throws them away. I'm strictly against littering, so I pick them up and use them for my projects. ;)

[@AndreasSpiess]

Interesting. Do they also come with a manual? ;-) Here, most of them are fixed that you cannot take them with you.

[@zvpunry1971]

Unfortunately no manual. And yes, there are also some devices that are mounted on lamp posts but they don't need a manual, they just have a giant LED display and show the speed. The newer ones don't show the speed anymore, only a blinking red ":‑(" sign. I always test them by speeding up, but they never show anything else.

[@AltMarc]

Let me guess, in which country you live….Switzerland ?

[@zvpunry1971]

Maybe the guy with the Swiss accent, but I have a different accent. Guess again. ;)

[@AltMarc]

"I think that there must be somebody who has too many of them…." I would say, you can't beat Switzerland, on how many of these devices (per km of roads) are used… they seem to be everywhere. But perhaps I'm wrong, and there is an even worst country?

[@athucpham5608]

Dear Mr. Andreas Spiess, I love your video, please help me make a laser speed gun. Thankkyou so much.

[@AndreasSpiess]

I only do videos, no consulting

[@Giblet535]

It's possible to use a pair of cheap camera modules and create a static array of Dict objects to map stereoscopic edge difference pairs to an estimated range. An ESP32 should be able to do this quickly in varying light levels, especially if a narrow portion of the cameras' images are used. Calculate the target's heading relative to the cameras, then calculate its speed based on heading and two range delta measurements. There are lots of existing code examples for edge detection in an image. Here's an ESP32 camera reading demo that can help with the data acquisition part: github.com/igrr/esp32-cam-demo

[@AndreasSpiess]

You are right. This principle can be used for short distances. I will show a device with this principle in one of the next videos. For longer distances, the method is less accurate (unless you extend the distance between the LED and the camera.

[@sofianfadli7910]

Andreas, you say that you doing research on your lab. Where is the money come from??? How do you get the money from what you're doing??? I'm curious. Thank you...

[@AndreasSpiess]

Maybe you watch my last channel update

[@obelix76]

Remember: Failure is *ALWAYS* an option :D (I wouldn't call the experiments in this video a failure. You just know now how to *not* build a laser speed gun :D) Maybe the range sensor can be tweaked. I'm not very well with those optical stuff, but maybe filtering the light that can enter the sensor (if there are no other filters in already) so only the color of the reflected laser is detected, might help. Or try a laser with more power (but mind the laws for Laser devices in your country... ;))

[@TuttleScott]

seems like any kind of filtering would interfere with its functioning because the speed of light changes in different materials.

[@AndreasSpiess]

As far as I can see the receiver diode/transistor has an IR filter applied.

[@HansVledder]

I love the hat! :-)

[@AndreasSpiess]

Mee too. I bought it at the last Harley Meeting in Austria... Maybe it will be used one day.

[@HansVledder]

Make sure to record that as well :-)

[@bijinapallitrinadh3002]

Hi help me with the code

[@AndreasSpiess]

There is no code because it does not work.

[@GRBtutorials]

LIDARs work using phase shift measurement, not actual ToF. Check this for more information: physics.stackexchange.com/questions/255947/how-do-laser-tape-measures-work.

[@AndreasSpiess]

You are right for rangefinders. But... Maybe you will watch my video on Sunday.

[@uwesoboszcyk8791]

Es ist so enttäuschend, wenn man feststellt, dass man Geld für ein Produkt ausgegeben hat, welches die angegebenen Daten nicht, oder nur unter unrealistischen Bedingungen einhält. Da verliert man schon Mal den Spaß daran, ein Projekt weiter zu verfolgen. Ich bewundere jeden, der dann sagt, "jetzt erst Recht" und einfach weiter macht. Bei mir fliegen solche Projekte oft in die Mülltonne. Allerdings hat man sich inzwischen auch daran gewöhnt, dass man den Angaben, insbesondere bei den billigen chinesischen Produkten, mit Vorsicht und Skepsis begegnen muss. Oft hilft schon ein Blick in das Datenblatt der verbauten ICs, um festzustellen, dass diese bei den für das gekaufte Platinchen angegebenen Daten am Limit oder auch darüber betrieben werden. Häufig auch zu sehen bei billigen Boost-Convertern.

[@AndreasSpiess]

Ich bin nicht bekannt dafür, dass ich schnell aufgebe. Sonst hätte ich kaum mit 61 diesen Kanal "gechschaffen" ;-) Mit diesem Projekt habe ich sehr viel über Distanzmessung mit Licht gelernt. Das ist vielleicht mal nützlich. Und wenn nicht, mindestens interessant für mich.

[@uwezimmermann5427]

These sensors with separate laser optics and receiving optics are most certainly not ToF-devices but using triangulation. The reflection of the modulated laser beam is recognized on a 1D or 2D camera sensor and then the distance is calculated using trigonometry.

[@AndreasSpiess]

I know of this possibility but thought, the base is very narrow to measure even 50 meters using this method. The short range LiDAR modules seem to use this method.

[@uwezimmermann5427]

in order to allow for the fast measurement at "short" (for lightspeed) distances, normally the laser and receiver in a ToF system should be mounted in the same package, as in the case of the new cheap ToF sensors available. But yes, after reading up a bit, there a version of ToF systems around which does not use pulses but a modulated laser diode where the analyzing electronics extracts the phase difference of the emitted and received, modulated signal.

[@AndreasSpiess]

You are right. After more investigations, I found that my device "modulates" the laser with different frequencies. In the end, it measures TOF, just not using time but using phase of a defined signal.

[@teeboytel]

has nobody noticed the video being 13:12 minutes long???

[@AndreasSpiess]

Most of my viewers watch it to the end.

[@publicmail2]

Prorate the distance into the refund plus 20%

[@AndreasSpiess]

:-))

[@dellyyop]

What's in your lab that we're not allowed to see?

[@AndreasSpiess]

Nothing I know of.

[@notofinterest]

Afaik the measure by triangulation not by time.

[@AndreasSpiess]

Triangulation is used for short distances. Since you speak German: kzbin.info/www/bejne/bHqpZoSQhNVla6c

[@notofinterest]

Andreas Spiess Gut, rednma deutsch... eine spannende sache, hab ich so nicht erwartet. Ganz glaub ichs immer noch nicht, aber gut...

[@johnlh100]

An excellent description of the principle of such laser distance-measurement devices can be found in kzbin.info/www/bejne/ppTaZIibaMh_i9E (in German). The principle is a tricky but simple interference measurement of the phase shift between the original and reflected signal. By the way, recently the highly reputable German CT magazine also issued an article on this topic, www.heise.de/ct/ausgabe/2018-5-Aufbau-und-Funktionsweise-von-Laser-Entfernungsmessgeraeten-3964045.html, only with the problem, that what they were describing was complete nonsense. They had to revise it later and I cannot remember that CT had ever been so wrong. Good to know, that also the best are making mistakes. I would expect (my guess only!) that laser speed guns use in principle similar interference-tricks for measuring speed directly by analyzing the small frequency shift caused by the Doppler effect as distance devices evaluate phase-shift for distance. The alternative of two subsequent measurements as used in your experiments has the major disadvantage, that the second distance measurement might not hit the moving object at the same location thus causing measuring errors that can be significant. Of course there are workarounds and compensation methods, but it’s also more elegant to get the right answers directly form the physical effect. Thanks for the nice video and you might give your device a second chance by designing a cheap automatic 3D-laserscanner, what could be a pretty demanding and usable project.

[@AndreasSpiess]

Thank you for your link. Elektronik recently even did an update video about the topic. I did some further investigations on my device and can confirm what he found. Long range devices probably work differently and really measure the time of flight. We will see when I get my new rangefinder. I was not aware of the CT article (even if I am a longtime subscriber of CT and Make).

[@gr7480]

Do you see live chat of bitluni? It s a great idea?

[@AndreasSpiess]

I saw that he made it. But I did not have time to watch.

[@pepper669]

Have you considered using the Lidar Lite V3 (especially the HP - high speed - version looks promising: buy.garmin.com/en-US/US/p/578152/pn/010-01722-10#)? Could you use its PWM signal output to measure the Doppler shift directly?

[@AndreasSpiess]

Looks like a very interesting product. I have a similar one here (for shorter distances) which I will test in the next weeks.

[@dominikwestner4707]

What technology are these displays using? www.speeddisplay.ch

[@AndreasSpiess]

The datasheet says: Messeinheit: Doppler-Radar Frequenz: 24,125 GHz

[@SidneyCritic]

Sounds like weak Laser, or sensor, or bad optics. I wonder if they can be boosted. Real world physical testing it always interesting.

[@AndreasSpiess]

I do not know if it is a bad design or if my device is defective. I have no comparison... I for sure will compare its results with the one on order (which is specified for 1500 meters !)

[@petermeiner3574]

I can't belive it, Andreas goes Steampunk!!?

[@AndreasSpiess]

Seeing is believing ;-) I love this handmade American hat.

[@TheTruthSentMe]

Their spec wasn't trustworthy, I guess.

[@AndreasSpiess]

Or I have a bad device. I do not know.

[@maicod]

pitty it is so weak

[@AndreasSpiess]

You are right!

[@truman7502]

Hello Sir, I am working in an Autonomous truck and I hope to connect with you to ask some questions about distance measurement technology. Is it possible to contact you via instagram or facebook messenger or email?

[@AndreasSpiess]

I share my knowledge on KZbin and unfortunately, have no time for consulting :-(

[@truman7502]

@@AndreasSpiess thank you for your reply