Autonomous robot's navigation

[EDV]

Since October 2007 I developed new object recognition algorithm "Associative Video Memory" (AVM).

Algorithm AVM uses a principle of multilevel decomposition of recognition matrices, it is steady against noise of the camera and well scaled, simply and quickly for training.

And now I want to introduce my experiment with robot navigation based on visual landmark beacons: "Follow me" and "Walking by gates".


Follow Me


Walking from p2 to p1 and back

I embodied both algorithms to Navigator plugin for using within RoboRealm software
So, you can try now to review my experiments with using AVM Navigator.

See also:


Navigation by map

[EDV]

AVM Navigator v0.7.4.2 update

Changes:

- The indication drawing was carried to ::Annotate method


- Into camera view was added 3D marker of target position of robot


...

See here about all other changes.

[EDV]

Fun with AVM Navigator



It's little demo of object recognition and learning from motion with helping of AVM Navigator.

All object rectangle coordinates are available in RoboRealm pipeline from external variables:
NV_ARR_OBJ_RECT_X - left-top corner X coordinate of recognized object
NV_ARR_OBJ_RECT_Y - left-top corner Y coordinate of recognized object
NV_ARR_OBJ_RECT_W - width of recognized object
NV_ARR_OBJ_RECT_H - height of recognized object

So you can use it in your VBScript program.

See here for more details.

[EDV]

Hi,

I noticed in the youtube video that you jiggle the object (your face or the book) slightly during the learrning phase. Is this an advantage?

Cheers,

Alan

In fact the AVM algorithm is not invariance to rotation and you should show the object for memorizing to AVM search tree under different angles during training for further correct recognition.

See also an example of using of Canny module as background for AVM Navigator: