Let's Make Robots!

Sharps, Sonar and a dead horse...

This is a dead horse, but...

Again, I am still sick of my sharp sensors. I am looking at  the LV-Max-Sonar(s) EZ(x) sensors as an upgrade... These are the "single-eye" sorta sonar sensors.

Open Question:

Has anyone played with these? Plus and minus for the output options (serial, an., pulse)? Is there clean data?  Etc. Etc.

Gimmie what you got.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
I can't say I've used em, but I've heard pretty good things about em. Great for the price, size and accuracy.
I can't help much either, but OddBot would be the one to talk to, I've seen several of his bots sporting one of these (SplatBot, Bot08M, etc).

Used 'em, like 'em, bought some more of 'em.

 I had a "better experience" with these than other sonars. Have a couple original EZ1s, and several more LV-EZ1s now. I usually use the analog output, but have used the PWM output too. The serial I'd heard was a littel inverted, but haven't worked with it. Heard some had problems with noise, but Maxbotix offers a solution. These have been very good at picking out obstacles and giving accurate info. Other devices I've used have been Polaroid sonars, Senscomp Mini-S, and the Devantech SRF04 and SRF05.

I'm not such a fan of the EZ-1s... even with the power filter I couldn't get them to be accurate enough for mapping (I reverted to using SRF-05 sensors and am happy with the results, but I think that's more of a single/double transducer thing than anything else). Actually, robotologist bought up my EZ-1s when I was getting rid of them, so I'm happy he's liked them.

I do like the EZ-3 for small robots navigating in close corridors (such as Trinity Fire Fighting). The narrow beam is something nobody else out there has. As long as you are going for a YES/NO there is something closer than X inches, they work great.

Regardless, I think it's important to use more than one type of sensor on bots. Sonar has a wide beam, IR is very narrow. Neither alone makes you (or your bot) entirely happy. A combination of sensors really is needed for robust navigation.

-Fergs

I found that the sharp IS sensor is more accurate - however that may be due to my running them off the same power pin with no filtering, as well as the lin. vs log. differences in the output.

 

check http://letsmakerobots.com/node/11215 for more information, and pictures!

Looking at your other thread and results, it's not so much that the sharp is more accurate, it's that it has a narrow beam of detection. The sonar has a wide beam, you can't just project the reading directly forward, the typical usage would be to use multiple readings together with an occupancy grid to model the environment. Single sonar readings can't be used for a mapping (hence why your sonar map is so  blotchy, it's picking up the walls).

-Fergs

The Sharps have been fine sensors too, no real problems in their use. It might be good to use a little math to straighten out the curve of voltage vs range, but the Sharps have done what their datasheet says essentially. 

Application might be to use sonars as a longer range area detector, with Sharps being used for shorter range point searching in areas your robot needs to pay attention to.

Range is definitely a difference, but I think the width of the detection beam is the biggest one. For instance, try following a wall with a sonar sensor, it's a lot harder than using an IR ranger, since you can't get an accurate read on how far away the wall is with the sonar. On the other hand, try avoiding a chair leg, if you aren't dead on, the IR misses it, but a sonar will typically pick it up.

-Fergs