Friday, June 14, 2013

XenonFire - A Tribute

Inspiration.
The world is run by this simple concept.  Inspiration always precedes innovation.  Would the modern automobile as we know it exist if Henry Ford had to re-invent the wheel?  What about the computer or modern medical science?  Isaac Newton put it best when he famously uttered:  “If I have seen further it is by standing on the shoulders of giants.”
When I first joined the NXTLog community, it was because of inspiration.  Seeing all of the incredible projects filled my mind with possibilities of what I could do.  Not just to admire and wish I had the pieces or the skills, but to use the ideas themselves as sources for my own constructions.
Therefore, inspiration was a major part of my robotics career even before I joined FIRST.  It is For Inspiration and Recognition of Science and Technology – the ultimate goal of the program is to inspire participants, mentors, and others about engineering technology.  And today, I bring you a tribute…to teamwork, coopertition and inspiration in general.
Those of you who have been around for a while might recognize these two creatures…in August 2010, NXTLog released the Robotic Pet building challenge.  Dogs, cats and birds headed the popularity list when scanning the list of entries…but most eyes paused in surprise when gazing at the two projects you see above.  CheshireCat150 and NatoNX had a slightly different take on the challenge.
So what’s so significant about two competing entries into the same challenge?  What’s the point?  And what does this have to do with the “inspiration” I was talking about earlier?  Well, in life and in FIRST, it is impossible to improve unless you are willing to learn from your competition.  Who knows, your competition might even be on your alliance someday…
Xenon (CC150’s entry, top) and BlazingFire (my entry, bottom) were two very different entries into the same challenge.  Both came at the True Problem of building a dragon in a different way, and the different Specifications each builder generated is evident in the Detailed Design exemplified in each one.  Xenon breathed fire and walked reliably, while BlazingFire raised his arm in anticipation of getting a treat.  What would I get, I wondered, if I tried to combine the two into one, and left the result as a tribute to inspiration and teamwork…?
This is XenonFire.  A week-long project attempting to combine the best of both Xenon and BlazingFire into a single design that, while not superior to either design alone, is a symbol of inspiration across the realm of robotics and life in general. 

LEGS/WALKING MECHANISM
The legs and walking mechanism are core to any quadruped dragon, and an essential part of the three you see here.  They put the LEGO in Motion, allow them Rumble over anything in their way, and Ultimately, make the Ascent wherever they want to go.   BlazingFire, while possessing a meticulous leg design, lacked the reliable, quick walking speed featured by Xenon.  My goal was to integrate a similar leg design and CC150’s elliptical-based advancement mechanism into a single dragon.  Turning would be accomplished by a central pivot, and the legs would be mechanically linked as to stay in synchronization at all times without sensor feedback and software aid.  To do this, I used a double-differential setup featuring two motors that allowed full power of both motors to be applied to both the walking and steering mechanisms at different times.  Each of the motors are connected to one of the outputs of each differential, except that one motor’s output is reversed by means of a gear train before entering the second differential (the one on top of the robot.)  This way, the bottom differential turns when the motors rotate in the same direction.  This drives the center shaft that drives the legs.  The top differential rotates when the motors turn in the opposite direction.  This drives the center turntable that allows XenonFire to pivot and turn.


NECK AND HEAD
The neck and head are important features of any dragon, and provide key clues if the dragon is friendly…or just hungry!  XenonFire, while having a head too small to feature a fire-breathing mechanism that Xenon sported (trust me, I tried), does have a little red tongue.  Horns are similar to his predecessor BlazingFire.

WINGS AND TAIL
Another feature I certainly wanted XenonFire to have was wings, and I believe you will see a very strong resemblance..!  The tail, while not specifically based off of BlazingFire, is also rather similar. 

CONCLUSION
XenonFire features a simple navigation program, utilizing the ultrasonic sensor on his stomach.  This Code allows the Orange creature to roam autonomously.  When he detects an object for more than 0.5 seconds, XenonFire stops, opens his mouth and roars, then backs away, turns, and proceeds on his wandering. 
Enjoy these other pictures of XenonFire I had some fun taking earlier today…
 
XenonFire is a tribute to that inspiration, to the collaboration that allows our world to function.  Special thanks to CC150 - for all the inspiration. 

Tuesday, July 17, 2012

Rough-Terrain Vehicles

First of all, both of the things you see in the picture are missing their NXT Intelligent bricks.  (I'm charging the batteries)

So, I decided to build a Pack-Bot, inspired by projects such as LNE and N-XTreme ATV.  (It's the one on the right)  The idea was that I could use power functions to operate the treads, and have a NXT minibot that came off the front (inspired by this project).  I decided not to attempt this complex undertaking, and I built a simple pack-bot instead.  I'm glad I did.  The gear ratio allowed for good torque and adequate speed, while the front flippers allowed the machine to be lifted over even the largest obstacles.  One of the only problem I ran into was that it could not turn itself back over when flipped onto its back - the front plate rubbed against the ground, and the flippers lacked the torque to "force" the robot back over.  To fix this, I added a single beam projecting upwards directly behind the flipper motor.  This allowed the packbot to like differently on its back, which reduced the load on the flippers.  The other problem is that the treads also slip on their sprokets when the robot is run very hard.  This might just be a result of the LEGO elements, but moving the flippers out one stud and adding a beam to the outside of the central tracks might help.

The thing on left is an exploration of the TECHNIC Unimog parts and pendulum suspension using a turntable.  It is a four wheel drive truck with front wheel wheel steering and rear pendulum suspension.  It is powered off two NXT motors, which power a 3-differential system that powers all four wheels.  This system was inspired by victor28859's projects.  All in all, I am very happy with the Unimog parts.  The 3-stud differentials and universal joints make it easier to make narrower models.  Also, no extra spacing is needed to comply with the TECHNIC system, which is based off of odd numbers.  The portal axles are very rigid and accept different gears.  The rectangular blocks allow for no slippage between the 3-stud differentials and the double bevel gears.  Anyway, this model was reasonably difficult to complete, as I repeatedly dissembled the body structure, continually seeing ways to make it stronger or more compact.  The only gear slippage I encountered in the drive train was between the 16-tooth gear and the 16-tooth differential gear.  I fixed this problem by switching the gear ratios in the portal axles to 3:1, which reduces strain on the drive train under the same load.

Now to see which one is better!


Friday, May 25, 2012

The Mobile Pneumatic Multi Arm Series

So, quite a long time ago, I decided to make a mobile arm.  Too simple?  I also wanted to use pneumatics.  Still simple?  I wanted to get more than three functions out of three motors.  So, MPMA was born.

All in all, it wasn't a bad start to the series.  The drive base was pretty quick, using a white clutch gear to drive forward and turn in reverse off of one motor.  The pneumatic switching mechanism, using a differential and inspired by this project, also worked quite well.  Even the compressor, operating off one RIS motor, worked pretty well.  However, I knew it was possible to drag even more functions out of three motors (MPMA derived 5) and I also toyed with the idea of using more than one motor to move the robot.  And thus was formed MPMA V2...



 MPMA V2 was built around the idea that an output from each of two aforementioned differential "multiplexor" units could drive a tread.  This would give the robot the ability to move forward with the power of two motors.  However, since the treads never rotated backward, MPMA V2 could not be maneuvered in even somewhat tight spaces.  The pumping unit was very slow (though this might have been due to a broken motor) and I never implemented the third pneumatic switch.  The machine's chassis was also weak, adding to the myriad of issues.  Leaning from this, I moved on...


 MPMA V3 was the best robot in the series so far.  It employed a much faster pumping mechanism and a two-jointed arm with a working claw.  The structure was not very strong, but it held together.  MPMA V3 also enjoyed better ground clearance than MPMA V1, and the NXT Intelligent Brick could be easily accessed and removed for battery change.  However, the second joint in the arm did nothing more than move the arm into a more nonfunctional range and was thus rarely used.  Also, the drive base could only go forward and turn in reverse, leading to the next member of the series...


MPMA V4 is probably the best robot in the MPMA series thus far.  Instead of clutch gears and differentials, MPMA uses a mechanical multiplexor custom-built by myself but inspired by this project.  Gaining six outputs from only three motors, MPMA has a fully functional drive base, a single-joint robot arm with working claw on a rotating turntable, and an on-board compressor.  Unlike its predecessors, only two of these functions can operate at the same time (drive base, switches, or rotation/compressor are the combinations) but they are fully functional.  I think that this is a good trade-off.  However, the mechanical multiplexor is prone to slipping the outputs are placed under too much load.  Also, the compressor is slow.

MPMA V4 will hopefully be entered into an unofficial NXT Robotic Arm building challenge.

Thursday, April 12, 2012

And, we're done.

Well, I finally finished the programming to an acceptable extent, pictured it, and published it on NXTLog.....here.

Enjoy!

Monday, December 19, 2011

Change Mechanism

I revised some parts of the program and added a section to calculate and dispense change.  However, there's still some issues to fix.  I'm also having some troubles with MyBlocks....

Sunday, December 11, 2011

Public

Well, this blog finally went public.

Enjoy!

Friday, December 2, 2011

One NXT

Mwuhahaha.

I made it run off of one NXT and five other motors off the PF battery box.  Sorry that I don't have any pictures of the two-NXT version.  I also apoligize for the terrible quality.
Let's see....

Off the NXT:
Motors
One NXT motor to analyze the size of the coin (inspired by a coin-analyzer I saw online)
One NXT motor to dump the analyzed coin into the sorting tray or down the return chute
One NXT motor to move the sorting tray
Sensors
One RIS light sensor to detect the coin in the coin slot
One touch sensor for the coin analyzer
One NXT light sensor to detect the coins coming out the change dispenser
One PF IR link to communicate with the PF

Off the Power Funcions:
Motors
One PF M motor for the change-dispensing mechanism
Two PF XL motors for the upper two dispensers
Two "newer" RIS motors for the lower dispensers

In order to use only one NXT, I had to eliminate a sensor so I could plug in the IR link in the freed port.  I took out the touch sensor used to detect the starting position of the sorting tray by replacing it with a chunk of code that saw when the motor was not able to rotate a set number of degrees in a certain amount of time.  This occurs when the sorting tray is in its start position and cannot go any further.  The motor cannot push further and the code detects that it is not making any more progress.  Thanks to other online project for this idea, such as this one.

Getting closer...hope to use this again to make money in December or onward.