Finished installing the Ultrasonic TX & RX on the head. I needed to bend one of the standard brackets to look the other-way so I could install TX & RX in a symmetrical fashion on the sides of the head. Wish Kondo manufactured these brackets in a Left & Right configuration. Oh well, if this was my only gripe...
Also finished installing 2 x IR sensors or as Kondo calls them PSDs. There were no specific instructions as to where to put them on the robot. Some KHR-2 based photos show them installed on the feet right above the sole. I chose to attach them to the arms. On the dummy servo right above the hands. Rationale for this decision was that I have more mobility on the arms so I can "look" around with the IR sensors with ease. Also when I open the arms to make the look like a T I get maximum distance possible between the IR sensors which will help in triangulating later on.
On the software front...
My love & hate relationship with HTH4 is thriving. After some struggle I managed to program an endless loop that listens to the sensors and throws a motion when an object gets closer than a threshold that I set. I was able to put a call to my GetUpConditional motion in there as well so now if the robot gets thrown on the floor it gets up itself without me triggering the motion. All cool & nice. The family pet dog hates #6 (yes that's the name of my robot) even more now since the Ultrasonic transmitter seems to bother her much!
Once above was accomplished I then progressed to expanding my autonomous behavior motion to be more selective. I managed to figure out how to chain to CmdAD boxes to form an If-Then-ElseIf-Then struct. Now the robot responds to left & right IR sensors being approached differently by raising the corresponding arm.
I wish part...
I really really really wish HTH4 had variables in it. I want to now be able to compare readings from 2 sensors and do something. Ideally I would love to see a CmdAD variant that has 2 drop-downs for 2 AD inputs and then allows me to do something based on a comparison operator. Failing this I wish there was a way to assign the current reading of an AD input to a variable and then do the same for the other AD input and then compare the 2 and go from there...
Unless I'm missing the obvious I think I am very close to depleting the autonomous behavior possibilities in HTH4... Robotis Bioloid is looking nicer & nicer as I read more about it.Statistics: Posted by ks_khr3 — Tue Jul 06, 2010 2:14 am
]]>