Asked — Edited

Scanning Room With Laser

Has anyone done a room scan with a laser or some other method?

What hardware would you recommend?

How much programming is involved? (I'm not afraid of programming, just curious how much of scanning task is handled by hardware)

What about cost? options?


ARC Pro

Upgrade to ARC Pro

ARC Pro is your passport to a world of endless possibilities in robot programming, waiting for you to explore.

PRO
Synthiam
#1  

There are some lidar units available, but i'm not a fan of laser. Mainly because it's a sensor that performs only one task - distance. I really like Kinect because it works as a camera also... this means object recognition, distance, vision, etc..

I have been working on a kinect interface - but slowly because I can't figure out what to do with it. It'll come to me:) I welcome suggestions!

If you have programming experience, the Kinect SDK is very easy to use and will compliment the EZ-SDK quite well

#2  

I am sure several of us can help with ideas of how to use Kinect.

Off the top of my head, say I want my Robot to go from my office to the dining room and not hit any furniture. I could have a general map of which rooms it needs to traverse, but it could use Kinect to identify obstacles and plot a course through each room on its own without the approach, retreat, turn, approach, retreat, turn action of just using the ping sensor.

Gesture control (you could do this with the camera too, particularly using something like Open CV, but I think the Kinect SDK may already have what you need built in).

Probably dozens of other ideas.

@mcbaker01, what was your idea for what a laser scanner would be used for?

Alan

#3  

Sounds like I'll have to google kinect interface.

I need something that scans a room and returns a multi-dimensional array. Are you familiar with JSON?

X should be the wall closest to north

Notice the north east corner, it's not a square room.

If the room is 6x6 I would expect this { dimensions: { x: 6, y: 6, degreesDeviationFromNorth: 3 }, map: [ [0,0,0,1,1,1], [0,0,0,1,1,1], [0,0,0,1,1,1], [0,0,0,0,0,0], [0,0,0,0,0,0], [0,0,0,0,0,0] ], readFrom: { x: 3, y: 1 } }

#4  

So are you using this just to create a map, or will the robot then interact in some way with the room and/or its contents, or are you using it so the robot can determine where it is?

In any event, I think Kinect would do just what you want.

here is some good info on what it does:

http://en.wikipedia.org/wiki/Kinect

Alan

#5  

The robot is going to traverse the room, do a survey.

Thanks,

I will check that out.

#6  

Kinect also has an array of 4 microphones. So you can use it for voice input. It also identifies the direction a sound comes from. The free SDK has a lot of sample code to get you going.

** Full Disclosue: I work on the Xbox Hardware team at Microsoft :-)

PRO
Synthiam
#7  

@mike, oooooooooh really?:) You're partially responsible for the developer edition Kinect that Microsoft sent me then! I spent a bit of time with the robot team from redmond. It was a good group of ppl to meet, we enjoyed our visit

#8  

Ah... this is great.

Hope you will be around when I have questions Mike;)