PDA

View Full Version : BRAT Based Mech?



Roboman
06-17-2009, 12:41 PM
Would a Lynxmotion BRAT (http://www.lynxmotion.com/Category.aspx?CategoryID=97) be capable of becoming a mech? In its stock configuration, it comes with 6 HS-425 servos, but if you buy 6 HS-645's, they offer a price cut, if you already own a BRAT. Since its so lightweight, it doesn't really require powerful servos, so I'm thinking the 645's would enable it to carry all of the extra weight from the gun, camera, etc.

lnxfergy
06-17-2009, 12:48 PM
The problem I see with such a mech, is that the guns and camera are gonna weigh quite a bit. Don't underestimate just how much payload it really is. On the BRAT, I'd be concerned you won't be able to shift the weight over, and that if you do shift the weight either the ankle servo would be overloaded or that you'd drag the feet badly (i.e. similar issues to Hagetaka...).

[edit to remove statements about 5990, since manufacturer insists HS-645s would be enough]

That said... you wouldn't have a terrible amount invested in a HS-645 brat mech if it didn't work... you could try it, but I'm just not sure you could carry that payload.

Edit to add: If you do go this route... be sure to switch to LiPo to help keep the weight down, you can easily cut your battery payload in half.

-Fergs

Adrenalynn
06-17-2009, 12:49 PM
Not even close to the payload required with 645's, I'm afraid. [Doh! Fergs beat me in there - with a better answer]

Roboman
06-17-2009, 12:50 PM
Oh well. I guess I'll have to look elsewhere.

lnxfergy
06-17-2009, 12:51 PM
Not even close to the payload required with 645's, I'm afraid.

Wow.. and people say I crush dreams...

-Fergs

lnxfergy
06-17-2009, 12:54 PM
Oh well. I guess I'll have to look elsewhere.

Manny's 4dof legs were pretty impressive. I doubt that he could keep that think walking for 12-15 minutes with the 645s... but with 5990s, I bet it'll be a competitor next year. His weight distribution was done very well, and if he can get his control solution to stop glitching (hmm, checksums and onboard controller), he'd be a heck of a biped to reckon with.

Connor's mech could be in the similar boat as manny, he just needs to scale his guns back a little and work on the weight distribution (he's using 6DOF for reference, so that's quite a bit more than a BRAT)

-Fergs

lnxfergy
06-17-2009, 12:58 PM
Then again... as I sit here avoiding doing work.... If the goal is a cheap mech, a quad is always gonna be less expensive. And here we go with me being a servo whore: you can't beat the bioloid system for cost. An AX-12 is $5 more than an HS-645MG, and for that you get twice the torque, thermal and overload shutdown, and a serially controllable, pose-and-capture-capable servo. Add a bioloid comprehensive frame kit for $99 bucks and you have more brackets and screws than you know what to do with (and way cheaper and lighter than any of the SES parts).

-Fergs

mannyr7
06-17-2009, 01:06 PM
Well said, Fergs. Don't even think about 645's carrying all that mech gear. You'll need 5990's and that's with a stripped down cam, light-weight guns (you won't be able to carry even one fully automatic Defender, ask Connor) and no pan and tilt for aiming. So, you'll have to create stances to aim with the body. And if you're not handicapped enough, you will deal with heat and power issues. For my bet, start with a Bioloid. You'll start off with torquier, more cost-effective servos that can take more power. Even then, Issy, a quad, had some heat issues, so you'll have to rest your servos and be a sitting duck. But I'm hardheaded and just think bipeds are kool!

-Manny

Wow, before I finished, there were 6 posts ahead of me!

lnxfergy
06-17-2009, 01:34 PM
Well said, Fergs. Don't even think about 645's carrying all that mech gear. You'll need 5990's and that's with a stripped down cam, light-weight guns (you won't be able to carry even one fully automatic Defender, ask Connor) and no pan and tilt for aiming. So, you'll have to create stances to aim with the body. And if you're not handicapped enough, you will deal with heat and power issues. For my bet, start with a Bioloid. You'll start off with torquier, more cost-effective servos that can take more power. Even then, Issy, a quad, had some heat issues, so you'll have to rest your servos and be a sitting duck. But I'm hardheaded and just think bipeds are kool!

-Manny

Wow, before I finished, there were 6 posts ahead of me!

All of issy's heat problems were solved when I went to 2dof. His servos never topped 65C on Sunday, and most of them were <50C (at default, thermal shutdown on an AX-12 sets in at about 100C I believe). Building a bioloid biped will be quite a bit tougher, the AX-12s are about half the torque of a 5990.... but I intend to build a bioloid biped for next year, it's just gonna take a bunch of engineering and magic....

-Fergs

tom_chang79
06-17-2009, 02:28 PM
I started with 645MGs on my Biped Scout (yeah, I know, it's much heavier then the Brat), and the only time it would stand up on its own strength was when I volted the servos to 7.2V (instead of the max 6.0V).

645MGs are powerful, but they are only 133 oz/in at 6.0V. Consider the 5990TG servos at 7.2V, it has over 400+ oz/in. Each 5990TG is capable of what takes three 645MGs to do...

Too much torque is not a problem usually, but too little torque means "gotta buy new servos"...

Guns, cameras, and other stuff do add up. The biggest contributor IMO are the batteries that need to fuel these additional electronics and electro mechanical devices, which will also add to the pay load...

lnxfergy
06-17-2009, 02:47 PM
Yeah. The biggest problem is not that the guns and camera weigh SO MUCH.. it's that a typical biped, with enough battery for 12-15 minutes runtime (and really, you want 15-20), is just really close to the max payload already... most of them aren't loaded up with servos such that they can have 300-500g of extra payload slapped right on with no effort.

Heck, look at hagetaka... $5000 in servos - but just too much payload/poor leg design, it couldn't work. It all is gonna come down to engineering lower-power leg designs and gaits in order to make a biped work well at lower budgets.

-Fergs

Roboman
06-18-2009, 10:59 AM
How about some of these (http://www.hobbypartz.com/4ptopromgmeg.html) servos? At 6 volts, they produce 200 oz.-in of torque, for $8.90 a piece!

Adrenalynn
06-18-2009, 11:31 AM
Have you heard the expression "get what you pay for"?

If you could get a biped to walk with those, I think you'd be the first person on the planet to do so. And _everyone_ has tried.

lnxfergy
06-18-2009, 12:11 PM
How about some of these (http://www.hobbypartz.com/4ptopromgmeg.html) servos? At 6 volts, they produce 200 oz.-in of torque, for $8.90 a piece!

I seem to recall Darkback having a real cool video of those lighting on fire...

-Fergs

DresnerRobotics
06-18-2009, 12:25 PM
How about some of these (http://www.hobbypartz.com/4ptopromgmeg.html) servos? At 6 volts, they produce 200 oz.-in of torque, for $8.90 a piece!

If I had a dime for every time someone thought these would be a good idea...

Roboman
06-18-2009, 12:41 PM
If I had a dime for every time someone thought these would be a good idea...

Ah... maybe not...:)

mannyr7
06-18-2009, 01:28 PM
There are no shortcuts and no free lunches. Save up, buy a 5990 a month, or some Dynamixels. Period.

Promagic
06-21-2009, 01:36 AM
I don't know exactly what you guys are carrying around that's so bad your $5,000 servos can't take it but I am going to make my brat work.

I have been working for months designing BB guns using different methods and I am getting close. I am going to win Mech-Warfare next year!
BTW.... Where's the videos???

Beware of the Halfling...

robologist
06-21-2009, 02:19 AM
beware of the halfling...

lol! :d

lnxfergy
06-21-2009, 08:56 AM
I don't know exactly what you guys are carrying around that's so bad your $5,000 servos can't take it but I am going to make my brat work.

I have been working for months designing BB guns using different methods and I am getting close. I am going to win Mech-Warfare next year!
BTW.... Where's the videos???

Beware of the Halfling...

Trendnet IP Cam = ~90g
Tank Gun = ~100g
100bbs = 12g
Hopper = probably 5-15g
AVRRA Scoring Board = 30g
Lexan 3x3" plate + sensor = 20g
pan and tilt servo (just about everyone wants them) = 60-100g
Armor, velcro, other incidentals all add up to more weight

So, estimated biped payload (from what we've seen, 2 tank guns, ~300 rounds, 2 scoring plates) = 416g. Add pan and tilt you are well over 500g. Most of these kits have almost no extra payload available. The bioloid kit humanoid (AX-12s have 2x the torque of an HS-645), has empirically been shown to have about 500g of payload, which is a pretty decent amount for a biped.

Oh.. and we never talked about extra battery power to run the cam, and servos under extra payload, for at least 15 minutes (and really, you want more like 20minutes of runtime so you can actually boot it up before the match starts, and not blow up you batteries if you dont leap through the netting to shut it off the minute the match ends).

A further consideration is that mounting 1lb of payload fairly high on a walker that does major weight shift is going to *really* stress ankle pivot servos, as well as probably have to shift that weight quite slowly for fear of falling over. That slow shift is going to further stress the ankle servos, and make said mech fairly slow. And bipeds are already pretty slow.

Not saying it can't be done.. but at least be aware of the challenges, and the physics behind them.

-Fergs

robologist
06-21-2009, 10:48 AM
(AX-12s have 2x the torque of an HS-645)

I'd thought that the torque rating was under a bit of dispute, referenced here:
http://forums.trossenrobotics.com/showpost.php?p=15298&postcount=177

The AX-12 appears to have more tight control and safety parameters, almost making regular servos seem like they are running open loop (not really, but no torque limits, thermal, or current safety) but torque doesn't really seem that different. The Bioloid system also seems to be more mechanically stable and perhaps places the lever points in a more advantageous position too.

lnxfergy
06-21-2009, 11:28 AM
I'd thought that the torque rating was under a bit of dispute, referenced here:
http://forums.trossenrobotics.com/showpost.php?p=15298&postcount=177

The AX-12 appears to have more tight control and safety parameters, almost making regular servos seem like they are running open loop (not really, but no torque limits, thermal, or current safety) but torque doesn't really seem that different. The Bioloid system also seems to be more mechanically stable and perhaps places the lever points in a more advantageous position too.

My understanding is that most of these ratings are probably just at the breaking point of the servo. The AX-12s could probably run with a higher torque threshold for shutdown... but you run the risk of breaking something. I would not recommend turning load error shutdown off completely... I broke the gear train of a servo on issy doing that. It's really too bad we can't get metal gears for the AX-12.

I think there is also another factor to look at here though -- I had more problems with heat than load on Issy. The AX-12s have a thermal shutdown (at a whopping 85C! *edit: thought it was higher... fixed that*). Under fully loaded with mech gear 3DOF Issy, I was seeing a shutdown from thermal overload in about 12 minutes on a few servos. Granted, a lot of that was a poorly designed gait -- but 15 minutes of hard work for a servo also incurs a problem with heat. Looked at a slightly different way, AX-12s run at much higher voltage, and thus lower current for the same power rating. Assuming both have a similar on-resistance for their H-bridges, an AX-12 should generate about half as much heat as a 5 or 6V servo.

Theres also that point where you just are over the top and things sprial out of control. All of the lift servos on 3DOF issy were running about 85C+... took that upper servo out of each leg (so about 200g of load), and servo temps all dropped to about 55C -- even after 18 minutes battling Gary, none were over 65C.

-Fergs

mannyr7
06-21-2009, 12:33 PM
http://www.youtube.com/watch?v=I8vY30_P8rk

darkback2
06-21-2009, 12:54 PM
Sweet Video Mannyr.

I guess that is almost proof positive that making a brat based mech is possible...I guess the key thing to remember is that there is no free lunch. Mannyr made a lot of really good choices such as mounting the batteries in the feet and using 5990s instead of hs 645s...but this mech is going to be almost as slow as mine in the long run.

A couple of things to keep in mind is that some of the faster quads are going to run circles around this biped. 7990s anyone?

DB

mannyr7
06-21-2009, 01:10 PM
Thanks DB! Yeah, it can be done. IMO, the BRAT is a better jumping off point than the Scout because of the reduced DOF. The chicken walker design is awesome, but in nature is reserved for creatures with hollow bones and disproportionately large thigh muscles. Those 3 extra servos per leg really add a lot of height and it's just basic physics. You're asking those ankle servos to push on the short end of an increasingly long lever. So, keep your biped as short as possible and any weight as far down as possible.

lnxfergy
06-21-2009, 03:55 PM
Thanks DB! Yeah, it can be done. IMO, the BRAT is a better jumping off point than the Scout because of the reduced DOF. The chicken walker design is awesome, but in nature is reserved for creatures with hollow bones and disproportionately large thigh muscles. Those 3 extra servos per leg really add a lot of height and it's just basic physics. You're asking those ankle servos to push on the short end of an increasingly long lever. So, keep your biped as short as possible and any weight as far down as possible.

"as far down as possible".. but no farther down! The problem of course with a short biped, is that you'll have really issues shifting the weight, you have throw the ankle servo farther to get the COG over a stable point -- which could really overload it.

-Fergs

mannyr7
06-21-2009, 04:12 PM
"as far down as possible".. but no farther down! The problem of course with a short biped, is that you'll have really issues shifting the weight, you have throw the ankle servo farther to get the COG over a stable point -- which could really overload it.

-Fergs
Actually, you bend the knee and thigh joints of the upward leg to increase ground clearance, while straightening the weight-bearing leg. Compressing the rising leg shifts the COG, reducing the amount of tilt you have to apply to the weight-bearing ankle. Then, rotate with the hips to step forward. At least that's the way I do it with 4DOF. I take a complete step with Clyde without tilting my opposing ankle more than 18 deg off-center. BTW, my ankle servos are 645MG running at 7.2V. So far, they run fine, but I plan to upgrade to 5990's when my finances recover from the Robogames weekend.:veryhappy:

lnxfergy
06-21-2009, 04:44 PM
Actually, you bend the knee and thigh joints of the upward leg to increase ground clearance, while straightening the weight-bearing leg. Compressing the rising leg shifts the COG, reducing the amount of tilt you have to apply to the weight-bearing ankle. Then, rotate with the hips to step forward. At least that's the way I do it with 4DOF. I take a complete step with Clyde without tilting my opposing ankle more than 18 deg off-center. BTW, my ankle servos are 645MG running at 7.2V. So far, they run fine, but I plan to upgrade to 5990's when my finances recover from the Robogames weekend.:veryhappy:

Interesting. How long have you had him walking in a given timeframe? Has he walked around for 15 minutes without resting? If so, how hot are the servos at that point?

-Fergs

gdubb2
06-21-2009, 05:05 PM
Looking good Manny..

Gary

mannyr7
06-21-2009, 07:10 PM
Interesting. How long have you had him walking in a given timeframe? Has he walked around for 15 minutes without resting? If so, how hot are the servos at that point?

-Fergs
Ten minutes before packs dump abruptly. No servo heat, except to the ankle 645 when balancing on one leg for an extended period while creating forward gait. Ni-Cads! grumble, grumble...

lnxfergy
06-21-2009, 07:55 PM
Ten minutes before packs dump abruptly. No servo heat, except to the ankle 645 when balancing on one leg for an extended period while creating forward gait. Ni-Cads! grumble, grumble...

This is very good news... maybe all my grumbling about bipeds is unfounded... I'm just hoping we can see all our mechs speed up a little bit, if only for the audience's sake.

-Fergs

DresnerRobotics
06-21-2009, 07:58 PM
Well, look at the advantages CLYDE has over the scout based robots we've seen. He had less payload due to only using the tank guns (instead of the Defender like Shadowscout did), and he had 4 less servos and accompanying brackets than a scout. Shorter legs = less torque needed on the DOF... and his batteries were in the feet, keeping a very low COG. All the bot had to be capable of doing was lift its feet and plant them, having the batteries in the feet adds a lot in the way of helping balance.

Adrenalynn
06-22-2009, 04:38 PM
OT Discussion Split - at multiple requests.

Now starts: http://forums.trossenrobotics.com/showthread.php?t=3335&page4 split ends http://forums.trossenrobotics.com/showthread.php?t=3335&page5

Robot Dude
07-09-2009, 03:17 PM
Would a Lynxmotion BRAT (http://www.lynxmotion.com/Category.aspx?CategoryID=97) be capable of becoming a mech? In its stock configuration, it comes with 6 HS-425 servos, but if you buy 6 HS-645's, they offer a price cut, if you already own a BRAT. Since its so lightweight, it doesn't really require powerful servos, so I'm thinking the 645's would enable it to carry all of the extra weight from the gun, camera, etc.

Hello everyone, Jim from Lynxmotion here. :veryhappy:

Ducks from thrown bottles and cans... :eek: Whew!

My son James has been working on a BRAT based Mech for about 3 days. He has it walking forward, backwards, turning in place left and right, a few other poses. Controlled from the Atom Pro and PS2 wireless controller. He can also rotate the turret, turn on a laser pointer and fire each of the two guns independantly. The turret weighs in at only 12 ounces and all he has left to add are the ammo hoppers. The BRAT has the 645 upgrade. In the video you wil see a little oscillation in the turret as it was first assembled with a 475 servo. He has since replaced it with a 645. The bot is surprisingly small, light 2lbs 12oz., and uncluttered even though it has almost everything required for battle. The video shows it walking at a leisurly pace but it can surely move faster. The feet do not lift completely off the floor in the video, but they could with some gait sequence speed changes. It's stable and reliable all powered from a single 6vdc 1600mAh NiMH pack. Not sure of the max run time, but it's easily 10 minutes. For longer run times there are more effecient battery chemistries available that could be used. So anyway, can a BRAT be used as the base for an RC controlled Mech? Absolutely! In fact the entire robot was built from off the shelf components from Lynxmotion excluding the tank guns which I bought from a place in Japan I think. Without further ado here is the link to the video...

YouTube - BRAT based RC Mech Warrior...

Here is a link to our forum where we are covering the details of the build. Not a lot of info there right now, and we were goofing around a bit, but there are some images and some more details.

http://www.lynxmotion.net/viewtopic.php?t=5224

Enjoy!

Connor
07-09-2009, 03:30 PM
Hello everyone, Jim from Lynxmotion here. :veryhappy:

Ducks from thrown bottles and cans... :eek: Whew!

My son James has been working on a BRAT based Mech for about 3 days. He has it walking forward, backwards, turning in place left and right, a few other poses. Controlled from the Atom Pro and PS2 wireless controller. He can also rotate the turret, turn on a laser pointer and fire each of the two guns independantly. The turret weighs in at only 12 ounces and all he has left to add are the ammo hoppers. The BRAT has the 645 upgrade. In the video you wil see a little oscillation in the turret as it was first assembled with a 475 servo. He has since replaced it with a 645. The bot is surprisingly small, light 2lbs 12oz., and uncluttered even though it has almost everything required for battle. The video shows it walking at a leisurly pace but it can surely move faster. The feet do not lift completely off the floor in the video, but they could with some gait sequence speed changes. It's stable and reliable all powered from a single 6vdc 1600mAh NiMH pack. Not sure of the max run time, but it's easily 10 minutes. For longer run times there are more effecient battery chemistries available that could be used. So anyway, can a BRAT be used as the base for an RC controlled Mech? Absolutely! In fact the entire robot was built from off the shelf components from Lynxmotion excluding the tank guns which I bought from a place in Japan I think. Without further ado here is the link to the video...

Enjoy!

Nice work Guys!!

A few things to note. That camera, won't work for MechWarfare.. Your going to have to go with a WiFi IP Camera, those mini spy cameras won't cut it at RoboGames or any other major event venue due to interfearance. You also have to keep in mind you still have to add the weight of the target plates (2 for a biped) and the target board with the Xbee on it. Also, don't forget the hoppers and the weight of the BB's.. and currently the matches are 15 minutes long. I don't believe that 645's will cut it either.. 5990's are the way to go, especially with the thermal shutdown protection. I speak from experience on this subject as being one of three people who built and attempted to compete with a biped this year.

Robot Dude
07-09-2009, 03:40 PM
I understand there is some additionl weight, but I still think the 645s will do fine. The servos are just not under very much load in this configuration. I knew the camera wasn't going to cut it, but we do have a more powerful version that should do nicely. Hey it's a start... And it's completely functional. :cool:

Adrenalynn
07-09-2009, 03:45 PM
No - it won't, I'm afraid. I wasn't able to use licensed 1wt professional 2.4 transmitters for the street-level video (300g, 12v, VideoComm Technologies). The digital 5.6/SS/FH transceivers (at 4lbs each) were juuust able to cut through the noise with 6dB antennas, but not enough to produce professional video - out came the 1000' cable spools and we hardwired it..

The spectrum analyzer told the story there. It's a mess. No off-the-shelf non-licensed video transmitter, especially in 2.4ghz, is going to cut it. You're going to have to go with something that has error correction, like an IP cam.

Robot Dude
07-09-2009, 04:03 PM
Yes I just got off the phone with Andy. He explained the large metal building with a bazillion 2.4ghz signals bouncing around... He's sending me links to a lightweight wifi system. So I'm on it. The video system will need to be improved, but I'm still confident the BRAT will make a great inexpensive Mech suitable for the competition without spening big bucks on servos.

lnxfergy
07-09-2009, 04:25 PM
I'm still a little concerned about payload:


Add wifi cam (~90g) + extra battery to handle 15 minutes of runtime + 15 minutes of camera (I'm guessing this is at least another 75-100g)
Ammo + hoppers, 200rounds = 24g + hoppers
Scoring, 2 plates + AVRRA board, if you strip extra headers, and remove the power circuit in lieu of having another 5V source you're looking at ~60g total for scoring

So then we have approximately 200-300g of additional payload (7-11oz). Once you've got all the payload on, and can show it walking and broadcasting video for 15 minute, we'll call it "completely functional".

-Fergs

ScuD
07-09-2009, 04:50 PM
Yes, there's work to be done...

But all I can say is, great work James!!

What happened to the little pats on the back here guys? I understand there's a lot more to the mechs than at first seem obvious, but if everyone goes the tried and proved way, we may as well just set a rule to buy this and that kit and attach this to there..

mannyr7
07-09-2009, 05:35 PM
But all I can say is, great work James!!

What happened to the little pats on the back here guys?

Lotta fun in a little package! But yeah, we also need a tilt for the guns to shoot "Towering Tyb's" four foot tall behemoth!

Again, great work!!

Adrenalynn
07-09-2009, 07:06 PM
I totally agree that it's a great start - and looks good to boot.

I think Jim gets a little less of a "pass" than "some new guy off the street" here, for understandable reasons - he doesn't need our "pass". He's been around and knows the score as well as anyone.

I wouldn't consider it offensive in his position - probably more of a compliment. :)

But yes - it's a great start and attractive - and pretty "off the shelf". I think it helps forward Mech Warfare and I'm glad to see he and his son involved. We need more involvement like his, so it's great he's stepping-up! I can't think of many better candidates to develop a turn-key starting point for Mech Warfare, and that will really open it up to the virtual masses.

lnxfergy
07-09-2009, 08:59 PM
Yes, there's work to be done...

But all I can say is, great work James!!

What happened to the little pats on the back here guys? I understand there's a lot more to the mechs than at first seem obvious, but if everyone goes the tried and proved way, we may as well just set a rule to buy this and that kit and attach this to there..

The problem I have, is when someone walks in touting a "completely functional" mech that isn't complete. Unfortunately, as has been experienced by many builders, a walker that walks just fine and dandy when nice and light, completely fails to work when you get all that mech gear on board (camera, guns, hoppers, scoring, and enough battery to walk around for 15 minutes). Posting a 30 second video of a biped carrying two guns, and a camera that won't work at RG, and calling it a "completely functional" mech is a dangerous proposition.

We've seen quite a few people express interest at low-cost mechs, but I'd hate to see someone run out and spend $300 on a mech, only to find out it won't work, and that they may have been more successful with upgraded servos and maybe a $500 mech (or whatever the costs end up being). I wouldn't normally jump down someone's throat, but in this case, I know Jim can handle it. I'd love to see a low-cost option, but before we tout it as such, let's make sure it really is an "option".

You will see innovation, I guarantee it, but it almost certainly will not be from someone who builds their first robot. There is a huge learning curve in robotics, as robots are composed of lots of technically challenging skills: mechanics, electronics and software. You can't hope to design a robust system right out the gate -- you just won't even know how much you don't know! There is a reason people sell robot kits -- they are an easy way to start. Beginners will have their hands full just assembling a kit, and programming it to work as a mech. They will learn a lot along the way. And then they will return in future years with innovative or new designs. Beginners really should stick to the proven road to learn what's going on, and then branch out from there. You wouldn't recommend that someone becomes a medical doctor by skipping the "tried and proved way" (i.e. med school), or that a person who never took a math class leap in and start teaching math at a high school, the reason being, it just doesnt work (99.999999&#37; of the time). The first step in learning, is to learn just how much you don't know, so that you can set an agenda for learning. Any researcher goes the proven way through the basics, and when they understand everything the system involves, they will be skilled enough to be innovative. The same should go for most roboticists.

/rant

-Fergs

ScuD
07-10-2009, 01:31 AM
You're right Fergs, but I still have some thoughts on that.

Look at Tyb's bot for example. In no means am I trying to put him down here, btw!! but I mean, if he hasn't got experience in robotics, who here does? He built a bot worth thousands of $$$'s, only to find out the ankle servo's couldn't handle it.

So on one hand that proves your point of the added weight and payload cannot be underestimated, on the other hand it also proves that spending a lot of money is definetely not a guaranteed way to get it to work (although this design is -a lot- more extreme than most bots)


Actually, the only point I was trying to make is, you can get people to accomplish a lot more with a little "congrats!, but..." then with "that's just not going to work.."

Motivation is key in this hobby, and who can be more enthousiastic about robots then us roboteers?

I must say though, you make a very good point on not wanting people to spend 300$ only to figure out it's water down the drain.

CogswellCogs
07-10-2009, 01:39 AM
Well said, ScuD.

lnxfergy
07-10-2009, 06:17 AM
Look at Tyb's bot for example. In no means am I trying to put him down here, btw!! but I mean, if he hasn't got experience in robotics, who here does? He built a bot worth thousands of $$$'s, only to find out the ankle servo's couldn't handle it.

I by no means was saying that you need to spend thousands of dollars... but there is clearly gonna be a minimum cost. Quite frankly, if you added up the cost of the actual parts used on the final mech, 2DOF Issy was probably the lowest cost mech at mech wars 2009.

You'll note, in my post above, the only thing I took aim at was the argument that this was a "completely functional" mech warrior -- which it is not yet! I did not say "that's not going to work" at any point in this thread, I have simply, and repeatedly, listed the components still needed to add to make a walker into a mech warrior. 99% of people who have actually built a mech have found that once they got all thier mech gear on there, walking performance was severly degraded or entirely gone.

I'm simply saying "get all that payload on there before you call it completely functional". It's way more payload than most people are putting on 2 and 4 leg walkers these days.

-Fergs

DresnerRobotics
07-10-2009, 03:07 PM
I think CLYDE, Jim's bot, and Issy are good indications that "budget" bots can be built. You certainly don't need to drop thousands into this competition to be competitive, and that is a great thing.

Keep in mind Jim has been building robots for 15+ years now, he's by no means a novice and I have good faith in him. Definitely deserves benefit of the doubt.

I'm thinking that given the additional payload that has to be added, we might be looking at 5645s for servos, however the BRAT design has the advantage of having much shorter legs (read: Less torque required) and lesser payload from fewer servos/frame.

ScuD
07-10-2009, 03:58 PM
@ Fergs: Apologies, you indeed only stated the bot was not fully functional.

I think the difference in opinion betwixt us lies in the fact that I'm not one of those persons who usally takes the known routes. For example, a person may take two weeks to explain to me why fire is hot, but I will not understand why or how until I've put my finger into it. It's an evolutionary thing, I'm not quite there yet ;-)

That's why I'm writing my own code for the Bioloid, for example. I could just as well download the libraries from the Robosavvy forums, but I chose not to, even though it'll take me a lot longer to get the robot running.

I've always felt that there is a chance you diminish one's creativity by teaching him things "from the books", but now i'm getting filosophical.

mannyr7
07-10-2009, 06:31 PM
I think CLYDE, Jim's bot, and Issy are good indications that "budget" bots can be built. You certainly don't need to drop thousands into this competition to be competitive, and that is a great thing.

Keep in mind Jim has been building robots for 15+ years now, he's by no means a novice and I have good faith in him. Definitely deserves benefit of the doubt.

I'm thinking that given the additional payload that has to be added, we might be looking at 5645s for servos, however the BRAT design has the advantage of having much shorter legs (read: Less torque required) and lesser payload from fewer servos/frame.

I think my recent video proves the BRAT framework could be used as a functional biped chassis. However, Jim is being as optimistic as I was, if he thinks Hitec 645's will move the BRAT, along with all the essential gear that mech-warfare requires. The feet are already dragging, more weight will only make it worse.

As for, "budget", I've long since passed the $1000 mark on CLYDE. Bear in mind, this is my first robot ever and am including the cost of research and testing of hardware, electrical, and weaponry dead-ends I explored along the way.:o
That said, I don't regret a thing. I've learned alot since first stumbling upon this forum in february and continue to learn something new daily. If my mistakes help some kid build his/her first walking robot then great! I'm an open book, ask away. I know that sense of pride when you bolt all these pieces of plastic, metal and wires together, push a button, and it actually does what you tell it to, instead of falling over in a twitching heap for the jillionth time.:cool:

Adrenalynn
07-10-2009, 06:53 PM
I know that sense of pride when you bolt all these pieces of plastic, metal and wires together, push a button, and it actually does what you tell it to, instead of falling over in a twitching heap for the jillionth time.:cool:

+Rep for just that part of the statement. Your reputation has to improve if you get a biped to do what you tell it to, sans "twitching heap". ;)

lnxfergy
07-10-2009, 07:14 PM
As for, "budget", I've long since passed the $1000 mark on CLYDE. Bear in mind, this is my first robot ever and am including the cost of research and testing of hardware, electrical, and weaponry dead-ends I explored along the way.:o

That's the reality of "research", you *will* spend 3-5 times as much as the final parts list (at a minimum!). While the final parts list to reproduce a 2DOF Issy is probably $600-700, I can't even tell you how much money actually went into development (for instance, I bought a full bioloid kit but 8 servos + a frame kit is far cheaper if a mech is all you plan to build, and I produced a full run of 25 arbotix boards to get the PCB for the controller).

A true budget bot will almost certainly have to copy a working design (which I hope there will eventually be several, including something brat based), or be built by someone who has already dumped loads of cash into robotics over the past years of learning and bot building. Learning ain't cheap in any hobby.

-Fergs

lnxfergy
07-10-2009, 07:20 PM
@ Fergs: Apologies, you indeed only stated the bot was not fully functional.

I think the difference in opinion betwixt us lies in the fact that I'm not one of those persons who usally takes the known routes. For example, a person may take two weeks to explain to me why fire is hot, but I will not understand why or how until I've put my finger into it. It's an evolutionary thing, I'm not quite there yet ;-)

That's why I'm writing my own code for the Bioloid, for example. I could just as well download the libraries from the Robosavvy forums, but I chose not to, even though it'll take me a lot longer to get the robot running.

I've always felt that there is a chance you diminish one's creativity by teaching him things "from the books", but now i'm getting filosophical.

No hard feelings Scud.

Hmm... I've gone the same route though, I decided I didn't like the complexity of the robosaavy version and just build my own AVR library for bioloid. I also neglected to actually read much on how to make robots walk and just played til it happened.

As for the "books", I forget 99.9% of things I read within a few hours (unless I do something related to it, or write it down...) so, I'm free of that diminished creativity! (of course, it makes remembering material for exams sorta difficult)

-Fergs

ScuD
07-11-2009, 08:29 AM
Hear, hear! That's exactly my problem, I can't remember stuff I read or was told, that's why I need to actually try stuff to see what/how/why it did / did not function.

I too didn't feel much for the complexity of the robosavvy versions, probably due to a lack of experience in C. I still debug my code in assembly... feel like a dinosaur.

Robot Dude
07-13-2009, 04:57 PM
Wow, I appreciate all the feedback. Yes I jumped the gun on the "completely functional" comment. I incorrectly assumed my high power video link would cut through the noise. Not being there I had no idea this was a problem. The good news here is the BRAT can walk and shoot guns quite comfortably with analog servos. We are still working on this. We just wanted to show the progress so far.

As for the 15 minutes of run time. That doesn't necessarily require another battery. The BRAT with 645 servos is drawing 300mA standing, 500mA walking! The guns take a 1 amp surge, but it's momentary. So with a good 1600mA battery it can walk for a good hour and a half. In fact James is still using the same pack for the bot and hasn't charged it since he started the project. We are ready to resort to LiPo if necessary though.

The bot is shuffling it's legs because that's what the code is telling it to do. It's a slow, stable, conservative gait. It's not like the servos can't lift the bot to balance on one leg. It certainly can, I've seen it in action. We just haven't made it lift the legs to walk as it's not necessary for this competition. It's not been optimised for speed but we will as we get further along.

I have just started researching the video problem. My first thoughts are to incorporate the video into a bidirectional control scheme, killing two birds with one microcontroller. Putting a Zigbee system in our DIY radio project would probably work. Now don't start quoting baud rates and throuput estimations, I'm ok with lower frame rates... It's just a direction I'd like to investigate.

So anyway, we will keep this thread up to date as we make progress. Thanks, Jim

lnxfergy
07-13-2009, 05:29 PM
Just a few notes before wasting a bunch of time. Max speed of an XBEE is about 230kbps, lets say your camera is 160x120, with 8-bit color depth, thats 153600bits per frame. Lets assume you can do 80&#37; compression of that, thats 30k-bits per frame... so thats 7frames per second. Assuming full data rate. Unfortunately, the resend rate is important in full transmission, if you are transmitting 100% of the time, you'll lose a lot of data. I found the mech tough to drive on 320x240, 160x120 is gonna be crazy hard I think.

Of course, it sounds like you want to use the same XBEE for up/down. That's really gonna kill your data rate, as you lose time each time you switch from RX to TX on the XBEE. I'd seriously steer clear of having high data density transmission on your control line, a glitching bot isn't gonna be fun for anyone (we did try at one point to see if I could run the mechDAR on teh same xbee as control, the end result was massive drop in data rate... Issy gets control input at 30hz from the PC, which is overkill, but only upstreams once every 5 seconds to report voltages/heats. It's that switch from RX to TX that really starts to slow stuff down)

-Fergs

Adrenalynn
07-13-2009, 06:19 PM
If you took the ntsc into an SA7111 to digitize it, then fed that into a hardware h.264 encoder, you could do 320x240 at 60kbit/sec for 7fps with forced intraframes. It'd look like crud [hideous DCT macroblock artifacts], but it could happen... I would want to sit very still and have my competitor sit very still when I was targeting though. ;)

mannyr7
07-13-2009, 06:52 PM
The bot is shuffling it's legs because that's what the code is telling it to do. It's a slow, stable, conservative gait. It's not like the servos can't lift the bot to balance on one leg. It certainly can, I've seen it in action. We just haven't made it lift the legs to walk as it's not necessary for this competition. It's not been optimised for speed but we will as we get further along.

The shuffling gait would be to your advantage, as no bb's would roll under the feet causing a "Scooby-Doo" type moment.

Robot Dude
07-14-2009, 12:52 PM
Anyone tried these? Little short on info, but it clearly say's wifi... Let me know. I may pic one up to play with. :D

Amazon.com: Securityman ClearCam Wi-Fi Interference-Free Mini Wireless Color Camera Kit with Audio: Camera & Photo

Adrenalynn
07-14-2009, 12:57 PM
I haven't tried a 404 broken link. What kind of stream does it put out? ;)

jes1510
07-14-2009, 01:21 PM
Is this the cam you are talking about?
Amazon.com: Securityman ClearCam Wi-Fi Interference-Free Mini Wireless Color Camera Kit with Audio: Camera & Photo

If so then it say s "WiFi interference Free". It's not a wifi cam, it's supposedly immune to interference to wifi.

Adrenalynn
07-14-2009, 01:24 PM
I was able to figure out from what the link was supposed to be what the camera is.

It's _not_ wifi. It's "wifi interference free". Because it's 900mhz.

We've talked about 900mhz before... Very short range, low quality video. Two channels total, so no multi-competitor events. Older digital cell phones will step on it. All analog cell phones will step on it. My wireless microphones will walk all over it since they pound its IF in the licensed band.

Specs to the camera: http://www.securitymaninc.com/product/clearcam2.html

This is like 2008 all over again. ;)

Robot Dude
07-14-2009, 01:48 PM
Yeah yeah, posted too soon. It was sent to me as a recomendation. I should have looked more closely before posting...

Adrenalynn
07-14-2009, 01:57 PM
My new new years resolution: Be nicer responding to Robot Dude's post. I don't think my smileys are cutting it, signifying my ribbing...

Robot Dude
07-14-2009, 02:45 PM
My new new years resolution: Be nicer responding to Robot Dude's post. I don't think my smileys are cutting it, signifying my ribbing...

Well I'm just trying to help the cause here... I was pretty disappointed at the turnout for the competition. James and I though a BRAT could do it nicely, and in three days time he had it walking, shooting, and at least short range video. The only thing missing was the scoring plates, which I haven't seen and don't know much about them. Aside from that it was completely functional (without prior intimate knowledge of the venue requirements). That alone was more than I saw any other biped do. We're planning to make some Mech Warrior specific add on kits for the BRAT, everything minus the guns. Again just trying to help the cause. So I have an interest in getting the video problem solved in as small a form factor as possible. I'm not sure why all the hostility. Peace!

Ducks again from all the thrown bottles and cans! :eek:

ScuD
07-14-2009, 02:50 PM
That's why I was a little ticked off, frankly.
It's the same problem with science, a lot of very smart people turn down theories by others because they "cannot be true", by which they may just be keeping off the next revolution in transport, power generation, ...

A forum is an arena of discussion, but if everyone keeps an open mind those discussions will lead to better solutions, instead of debates.

Power to the underdog! :D

Adrenalynn
07-14-2009, 02:56 PM
No hostility from moi. As I noted earlier (http://forums.trossenrobotics.com/showpost.php?p=32256&postcount=42) - you're certainly one of the few that could make it happen reasonably.

I think there's a somewhat jaded outlook because the bipeds proved to be tougher than assumed and expectations weren't really reasonable.

I've been sorely tempted to design a camera specifically for the event. As "gainlessly employed" as I seem to be these days, I need to watch the financial outlay... Digitizing NTSC, hardware encoding it, then flinging it over wifi isn't all that tough. In small quantity, a Phillips digitizer like the SAA7111 could feed an FPGA that did the encoding, then that feeds something like a strongarm that handles the packetizing and feed to the wifi chipset. The cheap cameras aren't made to be light weight at present, and have a lot of board and packaging and feature-set that we just don't need for Mech Wars.

I wonder if the Prop is fast enough to pull off a reasonable fascimile of at least software MPEG1? I have one sitting here... Hmmm... Simple Transport MPEG4 doesn't necessarily require a tremendous amount of flops if you lose the advanced motion prediction. Not the prettiest video in the world, but a wifi serial radio would then carry it. Hmmm.

ScuD
07-14-2009, 04:31 PM
I know zilch about video / video encoding, let alone compression, but I would definetely be interested in the "inner workings" if you'd ever find the time to come up with that!

Robot Dude
07-15-2009, 03:28 PM
James has added the ability to tilt the guns so an additional servo will not be required to shoot at higher mechs. ;) Here's a video. It can walk even with the guns not at center.

Will be adding one more video today, in a few minutes actually... (teaser) Naysayers are more than welcome, no, encouraged to attend! :genmad: thanks!

YouTube - Lynxmotion BRAT based RC Mech Warrior (2) Pan and Tilt

ScuD
07-15-2009, 03:39 PM
Hooray James!! Now get us that second movie allready! :P

Robot Dude
07-15-2009, 03:58 PM
Not wanting to wait till we sort out the video issues I wanted to put one question to bed now... Here is a BRAT with analog HS-645MG servos walking and shooting with 12 oz. of additional weight added to the highest point on the bot as a worst case scenario. Still stable! Still not drawing a lot of current! Still kicking a$$ and taking names! No kidding... The gait can be changed to improve the speed, but this is the normal stock gait from the tutorials on the website. OK? :cool:

YouTube - Lynxmotion BRAT based Mech Warrior (added 12 oz. to the top)

ScuD
07-15-2009, 04:00 PM
Great stuff guys, keep up the good work!!

jes1510
07-15-2009, 04:21 PM
I'm digging it!

Robot Dude
07-16-2009, 09:17 AM
@ ScuD

@ jes1510

Appreciate the good words. Thanks!

We are ordering the wifi camera of choice. We will even post a first person video when up and running. I don't expect this to be a real challenge. We'll be back! :D

jes1510
07-16-2009, 09:56 AM
Make sure you get the "w" version (TV-IP110W). The W designates the wireless. Trendnet makes a cam that looks exactly like the one you need without the wi-fi that has the same part number without the "w".

Robot Dude
07-16-2009, 10:07 AM
Make sure you get the "w" version (TV-IP110W). The W designates the wireless. Trendnet makes a cam that looks exactly like the one you need without the wi-fi that has the same part number without the "w".

Will do! thanks for the tip. :veryhappy:

SN96
07-23-2009, 11:35 AM
That 12 oz. payload demo was quite a myth buster. :cool:
________
NU50M (http://www.cyclechaos.com/wiki/Honda_NU50M)

SN96
07-23-2009, 02:25 PM
645's are fine depending on how heavy you want to go. My bratinator packs on a bit of weight with the arms, head, and grippers. I'm using 5645's and I dont think I would want to go any less with my situation. My bot takes true steps, so for a short period of time a good portion of the weight is riding on the ankle servo.

Here is the video of it:


http://www.youtube.com/watch?v=m1KkLdiCSRg
________
Drug rehab forums (http://www.rehab-forum.com/)

jes1510
07-24-2009, 12:50 PM
Got any updates with the Trendnet cam?

darkback2
07-25-2009, 10:50 AM
I've been looking at this thread for a while...Really digging the work your doing. Can't wait to see it fully functional!

DB

LTGunner
07-27-2009, 07:37 PM
Here's my first attempt at a BRAT Based Mech:


http://i67.photobucket.com/albums/h299/LTGunner/DSCF3441.jpg


Lots of work yet to accomplish; but it's a start...

Robot Dude
07-28-2009, 09:19 AM
Got any updates with the Trendnet cam?

Well when James got the BRAT working so well using the PS2 Controller we decided it would make a good kit on it's own. So a minor diversion while we make a new tutorial, then we will have something to show.

Robot Dude
07-28-2009, 09:23 AM
Here's my first attempt at a BRAT Based Mech:


http://i67.photobucket.com/albums/h299/LTGunner/DSCF3441.jpg



Lots of work yet to accomplish; but it's a start...


Nice looking bot! Looks like you widened the stance a bit, might make pivoting on one leg more difficult. Look forward to seeing what you can make it do. The BRAT just seems like a natural for this. 8)

Robot Dude
08-06-2009, 04:06 PM
Well when James got the BRAT working so well using the PS2 Controller we decided it would make a good kit on it's own. So a minor diversion while we make a new tutorial, then we will have something to show.

Well the new PS2 controlled BRAT is on the website now...
http://www.lynxmotion.com/Category.aspx?CategoryID=97


So we finally got the video of Hunchback with the wifi cam installed. Still don't have the ammo bins on there yet, but we're working on it! :D Enjoy...

YouTube - Lynxmotion BRAT based RC Mech Warrior with wifi camera!

jes1510
08-06-2009, 04:11 PM
Death to Cthulu! I love the vid man. If that thing can run for the full time and hold ammo then it looks like you have a valid platform. Congrats!

Robot Dude
08-06-2009, 04:16 PM
Remember this is with stock analog 645 servos. The performance would only get better using digital servos. 8)

DresnerRobotics
08-06-2009, 04:27 PM
Well the new PS2 controlled BRAT is on the website now...
http://www.lynxmotion.com/Category.aspx?CategoryID=97


So we finally got the video of Hunchback with the wifi cam installed. Still don't have the ammo bins on there yet, but we're working on it! :D Enjoy...

YouTube - Lynxmotion BRAT based RC Mech Warrior with wifi camera! (http://www.youtube.com/watch?v=3N5CJ_QPjQI)


/Applaud

Great work Jim!

Looks like I'll have to get some target boards in your direction. ;)

Robot Dude
08-06-2009, 04:38 PM
/Applaud

Great work Jim!

Looks like I'll have to get some target boards in your direction. ;)

I can't take all the credit. James did the bot and Beth and James did the video. I wish I had time to play... :D Yes we will need to borrow something to do the final video. Thanks! Jim

darkback2
08-07-2009, 12:16 AM
That is awesome. I especially like the fact that your using a shuffle gate because that means BBs won't be a problem. Can't wait to see it all cleaned up and trimmed out!

Great Job!

DB

nagmier
08-07-2009, 10:48 PM
Wow Great looking Vid Jim! Very nice work so far!

Robot Dude
08-31-2009, 02:48 PM
Beth and James finished a pretty good tutorial for the Brat Based Mach last week. Check it out.

http://www.lynxmotion.com/images/html/build156.htm (http://www.lynxmotion.com/images/html/build156.htm)

We will be adding content as we learn and add to the project. It's as complete as we can make it until we receive the hit/score modules. It's using the PS2 controller for now. The first diversion will be to make a custom controller based on an off the shelf largish PC based joystick. But the guts will be replaced with an Atom Pro with a Zigbee link to the bot. Enjoy! :cool:

sam
08-31-2009, 07:37 PM
Very nice tutorial! Congrats to all of you!

This might help some newcomers enter MechWarfare.

Sam

RevBryce
09-01-2009, 05:32 AM
I second that.


Very nice tutorial! Congrats to all of you!

This might help some newcomers enter MechWarfare.

nagmier
09-01-2009, 08:06 AM
Jim, Sparkfun carries a couple different Joysticks if your looking for something that doesn't need to be ripped out of an existing product.

http://www.sparkfun.com/commerce/product_info.php?products_id=9136

http://www.sparkfun.com/commerce/product_info.php?products_id=9182

Those may or may not be up to the task and they are pretty cheap.

Robot Dude
09-01-2009, 09:16 AM
Oh those are cool. :D What James is doing is a little different. He's using this joystick, which has all the buttons and stuff built in. He will be adding the Atom Pro / zigbee module to the inside of this joystick.


http://images.tigerdirect.com/skuimages/large/Saitek-AviatorJoystick-S261.jpg

RevBryce
09-01-2009, 05:23 PM
Nice stick!
could you fit a roboard in it for the video with HMD and not have a laptop at all?

Adrenalynn
09-01-2009, 07:25 PM
>> could you fit a roboard in it for the video with HMD and not have a laptop at all?

Sure, if you wanted to run something around 1fps and throw up on your own shoes, it'd be perfect. ;)

The real answer is: no. Not enough power for video.

RevBryce
09-01-2009, 09:32 PM
ok it was just an idea.
What would fit?

Adrenalynn
09-01-2009, 09:42 PM
You might squeeze a Nano ITX with 1.6Ghz Atom, but I doubt it.

Wait a few more years and you'll be golden.

[Mod Hat] Let's bring this back around to the BRAT now before I derail it any further...

DresnerRobotics
09-01-2009, 09:53 PM
Why a nano-ITX? Why not a pico-ITX?

Robot Dude
09-09-2009, 03:28 PM
OK! back on track. Here is some new content! What's missing from the photos is the camera battery, (just forgot to mount it) and the target boards. The area on top of the turret will house the xbee module for the targets. The PS2 receiver will be replaced with a similar xbee module for receiving commands. Kurt and James are working on the code now.

An image showing my paint job, the red wire looms, and the new camera.

http://www.lynxmotion.com/images/jpg/hunchie7.jpg
Here are some basic front, side, top views.


http://www.lynxmotion.com/images/jpg/hunch001.jpg

http://www.lynxmotion.com/images/jpg/hunch002.jpg

http://www.lynxmotion.com/images/jpg/hunch003.jpg

http://www.lynxmotion.com/images/jpg/hunch004.jpg

Adrenalynn
09-09-2009, 05:29 PM
It's danged purdy, for certain!

MSK Mech Commander
09-13-2009, 02:49 PM
Jim, Sparkfun carries a couple different Joysticks if your looking for something that doesn't need to be ripped out of an existing product.

http://www.sparkfun.com/commerce/product_info.php?products_id=9136

http://www.sparkfun.com/commerce/product_info.php?products_id=9182

Those may or may not be up to the task and they are pretty cheap.


Oh those are cool. :D What James is doing is a little different. He's using this joystick, which has all the buttons and stuff built in. He will be adding the Atom Pro / zigbee module to the inside of this joystick.


http://images.tigerdirect.com/skuimages/large/Saitek-AviatorJoystick-S261.jpg
Nice joystick!
I might get it...

Screw that!:p I'm gettin' this!:D!
Amazon.com: Saitek X52 Flight Control System: Electronics

http://forums.trossenrobotics.com/album.php?albumid=66&pictureid=327Just kidding, lol, maybe later...:p
http://forums.trossenrobotics.com/album.php?albumid=66&pictureid=327

robokoi
09-18-2009, 06:23 AM
Screw that!:p I'm gettin' this!:D!
Amazon.com: Saitek X52 Flight Control System: Electronics (http://www.amazon.com/Saitek-X52-Flight-Control-System/dp/B00030GSJY/ref=sr_1_5?ie=UTF8&s=pc&qid=1252871468&sr=1-5)
http://forums.trossenrobotics.com/album.php?albumid=66&pictureid=327

that's actually a beautiful control pair. was thinking about it myself since we use it for work with a lunar simulator and other vehicle controls