Society of Robots - Robot Forum

Electronics => Electronics => Topic started by: airman00 on January 27, 2008, 09:16:35 AM

Title: Infrared vs. Ultrasonic
Post by: airman00 on January 27, 2008, 09:16:35 AM
I was bored today so I decided on writing a tutorial on ultrasonic and infrared sensors.

So many people have asked which sensor to use , so I wrote a tutorial to answer that question

http://www.societyofrobots.com/member_tutorials/node/71 (http://www.societyofrobots.com/member_tutorials/node/71)

If I need to add anything or made a mistake please tell me.
Title: Re: Infrared vs. Ultrasonic
Post by: Steel_monkey on January 27, 2008, 11:40:06 AM
I was also going to write article about sonar, because I was going to make my own (but haven`t yet). I attach some schematics and datasheets for additional explanation about how all this works and for anybody who wants to make his own ultrasonic rangefinder. 
h*t*://rapidshare.com/files/87071798/ultrasonic.zip.html

P.S. Please, stick this topic and don`t chat here.
Title: Re: Infrared vs. Ultrasonic
Post by: Admin on January 27, 2008, 08:09:44 PM
Quote
Infrared is cheap( $10 -$20) and not very accurate , ultrasonic is expensive( $20+ ) and more accurate.
This isn't really true. It greatly depends on how its used, why its used, the environment, etc. Sharp IR has much better pin point accuracy than sonar.

You should talk more about the strengths and weaknesses of each sensor, perhaps how the sensors can work together to make up for each others faults.

For example, in the land of pillows, sharp IR wins. In the land of bright sunlight, sonar wins.

Oh for both of your tutorials ,your commas ,periods , and parentheses are all in odd locations (,you should check on that   .) :P
Title: Re: Infrared vs. Ultrasonic
Post by: airman00 on January 27, 2008, 08:18:38 PM
Quote
Infrared is cheap( $10 -$20) and not very accurate , ultrasonic is expensive( $20+ ) and more accurate.
This isn't really true. It greatly depends on how its used, why its used, the environment, etc. Sharp IR has much better pin point accuracy than sonar.

You should talk more about the strengths and weaknesses of each sensor, perhaps how the sensors can work together to make up for each others faults.

For example, in the land of pillows, sharp IR wins. In the land of bright sunlight, sonar wins.

Oh for both of your tutorials ,your commas ,periods , and parentheses are all in odd locations (,you should check on that   .) :P


you got it admin, ill do that this week. About all those periods and commas, very wierd i did not notice typing all of that in . I need to fix that !

P.S. I'm going to add the land of pillows thing to it!
Title: Re: Infrared vs. Ultrasonic
Post by: benji on January 28, 2008, 08:56:55 AM
well , nice tut but ... you kind of trashed the sharp IR next to the sonar, i have a different point of view,,,
you have not mentioned that the ultra-sonic has a beam so wide that makes it a stupid sensor in many many bot applications(30 degrees).
the only thing sonars are above sharp irs is the long range ..
Title: Re: Infrared vs. Ultrasonic
Post by: airman00 on January 28, 2008, 02:56:31 PM
well , nice tut but ... you kind of trashed the sharp IR next to the sonar, i have a different point of view,,,
you have not mentioned that the ultra-sonic has a beam so wide that makes it a stupid sensor in many many bot applications(30 degrees).
the only thing sonars are above sharp irs is the long range ..

true , thats why i need your  input!

i am biased towards sonar since I have never needed the thin beam of the IR , and for all my applications sonar was the best. I will add this in when i get teh chance.
Title: Re: accuracy Infrared vs. Ultrasonic
Post by: Armouralis on January 08, 2009, 06:45:12 PM
hey all,

what is the EXACT rated accuracy for the infrared or ultrasonic sensor?

like if i tell a robot to stop 3 inches away from an object, then what is the tolerance for each type of sensor for example if i am using the exact same ones that admin used on Fuzzy,the omni wheeled robot

is it like 1/2 an inch, 1/4 inch ect.  i did a google search and i checked the datasheets but with no real result....


thx in advance =P
Title: Re: Infrared vs. Ultrasonic
Post by: Webbot on January 08, 2009, 07:08:05 PM
Wow, old topic!

Distance is one thing. Sonars have a pretty useful distance range - whereas most IR detectors, for the same range, require quite large bursts of current. But the main difference is that IR devices send out a needle thin beam of light whereas sonars send out a cone of noise. So IRs may be useful to measure the gap between two objects to decide if the robot can get through the gap whereas the sonar would just see it as one big barrier. For the same reason: IR sensors dont see small objects (unless pointed straight at them) such as chair legs - whereas sonars are much better.

So sonars give you a broader/fuzzier view of the world. In simple apps this makes it easier to avoid stuff as they are more forgiving. But if your app needs to get the robot through a gap which is fractionally wider than the robot then an IR may be better.
Title: Re: Infrared vs. Ultrasonic
Post by: Webbot on January 08, 2009, 07:11:43 PM
Sorry for dbl post but just realised I didn't really answer you question.

Both devices should be capable of measuring to say 1/2 cm. Although the Sharp IR devices do generate some 'random' spikes (search on other posts for the Sharp detectors). But, as per my previous post, then there are considerations in terms of how fuzzy you want the 'width' of the eye to be at that distance.
Title: Re: Infrared vs. Ultrasonic
Post by: Admin on January 10, 2009, 09:21:01 AM
To be honest, I'm still shocked that many sensor datasheets don't tell you the most basic things such as accuracy. Only until recently are the Sharp IR sensors coming with datasheets at all!

That being said, if the datasheet doesn't say, the only way you can know is buy it and test it under 'realistic conditions'.

There are also modifications you can make to your sensors to make them more accurate, as well as more intelligent algorithms you can use.