Let's Make Robots!

Accurate distance sensor


Total newbie with regards to robotos, but fascinated by the topic. I'm working on a hobby project, which is not really a robot, but I need "robot" components. I'm searching for an accurate distance sensor, capable of sensing distance between 0 and 5 cm (could also between 1 and 6, for instance), with accuracy around 0,1mm. I see some of you are using Sharp distance sensor GP2Y0A02YK0F and similar, but I have not found what is the accuracy of those sensors, so I don't know if it would be appropriate.

Any pointers on how to achieve that will be greatly appreciated. Also, I don't need fast measurments. Measurment every second or two seconds is enough.

Kind regards,


Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

This came to my inbox today.  It will get you "sub-micron non-contact" accuracy if you can figure out how to use it.


Thanks! I'm looking into this...

I appreciate it.



Perhaps if you told us what it is you're actually trying to build/achieve here we could present you with a better solution than one which requires precision beyond that of the Hubble Telescope optical fix. Be as direct, detailed and specific as possible. Don't worry if the request isn't for a "robot." Most robots don't look like robots after all, and we sometimes argue over whether things that clearly look like robots really are, or if they're just remote-controlled cars with arms (or toothbrush heads with pager motors glued to them for that matter.)


sorry for not being too clear about it. It is not a robot, not in a classical sense, it's is more of an instrument. We play with RC cars (1/10 on-road racing cars) and we set-up those cars for each track and after each run. The height changes a lot (using foam tyres) and I'd like to build a "height gauge" - to measure front and rear height of the car on a "board". So the idea is that this board would have embedded two distance sensors with which I could sense the height of the object (front & rear) and display this on a simple LCD on the board. And also measure the weight, but that's not such a big problem. That's it.

So, in order for the measurment to be useful, it need to be accurate to about 0,1 mm (0.5mm would actually work too). Hope that clarifies the goal a bit and that it's not too cryptic anymore.

No it's not a robot but you're in a good place for help with this kind of thing.
The first thing that occurs to me is that a traditional distance sensor of any kind is probably not what you're looking for here. The problem with using a distance sensor to measure height of an object is most of the distance you seek to measure is obscured by the volume of the object itself (ie, from the board level to the tops of the tires or hood/boot of the car there is mass and surfaces that a distance sensor would not penetrate to give you an accurate measurement. You might be able to "hang" sensors from above the board and get a subtractive measurement, but there will be some real challenges in precision mounting and you probably still can't expect results with precision to half a mic with a commercially available non-contact sensor.
You might consult some of our real engineers (BDK, Dan, and jscottb, among others) but my gut feeling is that to achieve what you're describing will involve either a physical contact method (like calipers) or a vision processing method.

I was just thinking something mechanical, some bizarre contraptions in my head which probably wouldnt work so I would go with operation: real engineers :P


what is the precision (accuracy) one can get from the Sharp sensor? What about those sensors with PSD chips in them, that detect distance by reflected angle (or something along those lines)? I guess they're not available as a hobby packages.


I am not sure on the resolution of a sharp sensor as I dont use them (I use the HCSR04 ultrasonic rangefinder which certainly wouldnt be capable of measuring 0-5cm with 0.1mm precision, I'd be cautious trying it with 5cm accuracy to be honest), I dont think they actually return a linear result actually. DanM I think had a very good writeup on them as do others I presume.


If you want to try distance by reflected angle (which actually the sharp is probably doing anyway) you can look into LIDAR. Wiicams are useful for that actually. I am not sure on the precision of these either. If I recall it might be Gareth that had the writeup on LIDAR although there are probably some good ones via google. Oddbot may have also made one.



EDIT: GroG's is the one I was looking for not gareth: http://letsmakerobots.com/node/24259 He doesnt mention the resolution/accuracy/precision particularly so that would be a case of having to invest and try it I think which might not be an option.

Rocketbrand studios sell the sensor from a wii remote (its i2c).


I'll check the mentioned resources.