DPRG
DPRG List  



DPRG: RC / Hexapod / Video - Progress Update

Subject: DPRG: RC / Hexapod / Video - Progress Update
From: Jim Frye jfrye at lynxmotion.com
Date: Tue Jan 13 19:18:14 CST 1998

Hello all,

New progress report of my personal hexapod project.
>
>Goal: 
> Build a small 3 servo remote piloted hexapod.
>
>Features:
> Standard Lynxmotion Hexapod Walker (with camoflage paint of course).
> Futaba 4 channel RC set (left stick = motion control) (right stick = CCD
>camera)
> Full pan and tilt of CCD camera (B/W)
> PIC processor to convert RC signals to stick position TTL for basic stamp.
> Basic Stamp control for walking gait.
> Video / audio transmitter on board.
> LCD "TV" for monitoring walking progress.
>
>Progress:
> So far I have the bug built, 3 standard servos installed, pan and tilt is
>built from 2 micro servos, basic stamp is wired and installed, RC receiver
>is installed, PIC processor to intercept the left stick pulses is prototyped
>and installed. The CCD camera is not installed yet, The LCD display has not
>been purchased yet. I have a video / audio transmitter that runs on 9V but
>it needs to be put into a smaller enclosure, and is not installed. 
>
>Sucesses:
> The thing will walk under RC control and the pan and tilt works great. The
>robot weighs in a 1.2 lbs now with a bit more circuitry to add, the weight
>should not be a concern. I have tried sevel Futaba RC sets with great
>results. I will be trying a Hitec set when it arives. 
>
>Failures / Fixes:
> The thing is sucking the nicads dry real quick. I have measured up to 800mA
>spikes when walking. This is not acceptable so I am trying to optimize the
>setup to reduce the current consumption. I have adjusted the servos for the
>legs to maximum throw for the required leg throw and lift to give the max
>torque which resulted in some pushrods bending under the large load. I have
>ordered hobby ball joints and will replace the wire pushrods. I have found
>the servos have a greater torque at 6v rather than 4.8v. So I am putting
>together a 5 cell AA pack to replace the 4 cell pack. I may need to go to
>the Nickel-Metal Hydride (1200mAH) but am not anxious to spend $29.00 for a
>small battery. 

I have received the ball joints from Great Planes and the 1/16" music wire.
Took a few hours to change over. The robot now has much better torque
transfer from the servos to the legs. After putting together a 5 cell (6
volt) AA hump pack the search for the large heatshrink was under way. I
found a local hobby shop had some that fit nicely. The guy says it's $1.00
per foot expecting a complaint because I only needed a few inches. I said
give me 10 feet... I like bulk. :) Battery pack assembled and charging. The
robot now walks fine with no slop or moan from the servos. Strangely enough
the most dramatic help was when I changed the gait from a slower graceful
gait (providing intermediate steps for the servos in for next loops) to a
more bold gait where the positions are slammed. It is only a bit more jerky
but much faster and stronger. It is funny how these little bots take on a
personality. A bit of code tweaking here and there now the robots vertical
leg lift returns to mid position when told to stop walking, cool!. The
current consumption now spikes in the 500 mA region but quickly is reduced
to 30 mA when not stepping, also cool! I estimate quite a run time with 600
mAH cells.

>Future:
> I will be ordering the CCD camera soon and installing the whole mess onto
>the robot in the next few weeks. I antisapate a simular problem with power
>for the 9v side as it powers the basic stamp 2mA, PIC 2mA, CCD camera 20mA,
>vid / aud transmitter ~50mA I also expect the video transmitter range to be
>a dissapointing 100 ft. 

No CCD camera yet. I am considering one for $49.00 from Marlin P. Jones and
a couple from Ramsey for around $100.00 each. Just don't know much about
these dudes to start throwing money. Like how does a CCD with a low res.
sync to a TV with a fixed res? Anyone in the know chime in. I tested the
video transmitter I am planning on using with my camcorder video output. It
works fine. Will it work with a 300H 240V resolution CCD cameras composite
output?

>Closing:
> I don't expect this to be a great performer... I hope it suprises me. It
>will be somewhat slow (a few inches / sec.) and the operating time will be
>only 1 to 2 hours. However it is a great learning project for me and I am
>enjoying the oppertunity to overcome some huge obstacles. DO give me any
>feedback you may think is appropriate. Lets get the thread back on track and
>talk about real robots for a change. :) Thanks, Jim

It is definetly starting to suprise me. I promise to have images available
soon. I have been extreamly busy lately working on new stuff. Thanks, Jim


Jim Frye                             ---------
Lynxmotion, Inc.                   /   (o\ (o\)----
104 Partridge Road                +-----\ \-\ \   o \
Pekin, IL 61554-1403 USA        GET A   |\o) \o)     |--------
Tel: 309-382-1816               GRIP!   |/o) /o)     |--------
Fax: 309-382-1254                 +-----/ /-/ /   o /
jfrye at lynxmotion.com               \   (o/ (o/)----
http://www.lynxmotion.com            ---------

------------------------------

More information about the DPRG mailing list