TurtleBot3 arrived and well, I couldn't resist

 I originally ordered the RPi4 version; however I was contacted by Sparkfun.com to say that was not in stock and there was a min. of 2 weeks lead time due to worldwide delivery delays etc...

Initially, I thought, "tch! typical!", then the nice support person asked if I wanted to use an RPI3 version instead, as they had 1 of those in stock and basically the kit is exactly the same, it's just a different RPi.

Well....I already have an RPi4 inside the Robot Dog/Cat from Freenove....so, I could just swap that out if I really needed the performance....so, I agreed (with a little discount ;-) ) and well, the TurtleBot3 arrived the next day.

Here's some images of me being on conference calls and making the TurtleBot3 on my home/work desk - it was pretty simple/easy - however, my only observation was the lack of clear instructions that should show putting little nuts into the bottom/insert of the PCB board holders, once I figured that out all the PCB boards were then nice and stable.... however, there were a few of the metal supports that were not a tight fit, the screws were perhaps a little too long and gave a little gap so the support strut could spin/move - fortunately most were okay.  I was wondering if this was by design to make the whole structure less rigid.  We'll see whether it makes any difference.

I had already downloaded and setup the 32Gb SD Card for the RPi3 - however, the HDMI-to-VGA adapters that I need are on my work/work desk! So, I'll now have to wait until next week before I can start dipping into the ROS coding side of things - which is not so bad, as this was meant to be a "play toy" for my week off work starting Oct 18th....

Building out level 1.....yes, I originally put the battery the wrong way around! I figured that out later on once I attempted to plug the power cable in:

The LIDAR is securely connected to the RPi3 (via a little adapter) and the OpenCR PCB is connected to control everything, the battery is all charged up and ready to go.... however it will have to sit on my desk for a little while longer, teasing me....

I actually have high hopes for this little bot. Why is that?

Well...I want to apply some of my newly acquired Data Science / AI techniques into it - mainly the usage of 'Deep convolutional Q learning' along with some 'A3C intuition' and some 'LTSM' (Long Term Short Term memory) - so that I can start to have some PROPER SMARTNESS going on inside the brain of the robot.

What I DO NOT want is just a little robot that navigates around on its own - I have many of those.  What I want is a robot that has a memory & a brain - "I have not seen that before, what is it? let me connect to Google Images search and see, let me compare it to my CNN image model, let me then get some attributes about it, what relationships does that object have with anything else? how can/will that affect me if I perform any actions in relation to the object?... ie, that is a fireplace, that has a roaring fire within it..."

You get the idea.  Let's start doing some limited, but, proper AI....