I have been watching the Bioloid stuff. It’s relatively new so I don’t yet know if there are any gotchas. The servos are plastic gear, so there may be issues there. I don’t know much about the programming either. Not sure if it’s a very powerful programming environment. I’m not overly enthusiastic about the daisy chaining of servos, because the wires to the first servo have the burden of carrying the current for all of the downstream servos. The wire doesn’t look heavy enough to do this. I’ve seen some of the videos and well lets just say I was not impressed with the movements of the robots.
It’s funny to me that the one place that made no bones about criticizing my use of plastics in robotics jumped on board with a totally plastic construction set. LOL I am in contact with Robotis and will likely bring the Bioloid stuff to our site for sale. We may wait till Jan 07, before we do it.
Hitec was supposed to offer the bidirectional servo control scheme to us a long time ago. I basically just got tired of inquiring about it.
I am interested in hearing opinions of others on the forum, c’mon, don’t be shy… Should I bring this stuff in?
Those are reasonable concerns, though I’ve been casually following some of the Bioloid forums on the net, and haven’t heard of anyone having problems with them.
I don’t know much about it either, except that it almost certainly runs under Windows which means I won’t be using it anyway. I’m going to end up writing my own sequencing and control software pretty much no matter what hardware I use. But I realize that others may not be in the same boat.
That’s an interesting point — though again, I haven’t heard of it causing any problems yet. But as you say, it’s new. It could be that some of these issues will be coming to light soon.
Did you watch this one? Be patient and wait for all the silly posing to be over — at the end, it walks (and avoids running into things), forwards, backwards, and turns. It seems like a good walk to me.
There are some users at the RoboSavvy forums trying to work out the motion sequence for doing a cartwheel. We’ll see how that goes.
That’s good news! In addition to the kits, please try to carry the individual components as well, as much as possible. It helps folks like me to build up a useful collection over time.
But I too am interested in hearing what others have to say.
Oh yes, I just thought of one other advantage to the Hitec/SES approach… you can get decent servos (strong enough for small bots) for $15-$20. This means a lower entry barrier than with Robotis, where $45 is the minimum.
(Of course, the $45 AX-12 servo comes with the equivalent of a C bracket and a multipurpose bracket, which helps a bit.)
I have started keeping track of the Bioloid stuff also, but I am more interested to know if their bracket hole patterns are compatible with the SES brackets.
I am going to do some more research and see what I can find out. I will pass whatever I learn on to you.
I don’t think a daisy chain would be bad if it were only 3 servos or some other small number of servos. I have not seen how big the connecting wire gauge is, so I can’t say much about this right now but trust your judgement. Would the wire handle he current for 3 servos?
I do know the Open Servo folks are having noise issues with i2c lines being too close to power and ground, but I think that has more to do with their pin configuration as it does not follow the recommended standard. They are changing to conform though, so this is good. I want to use Open Servos in my Walk 'N Roll bot.
You are so picky! I like that about you though.
This is interesting, to be sure. Hmmmm, I wonder what they were thinking. I won’t doubt your judgement on plastics anymore. You have good stuff.
Well, I am interested, but price is a real issue for me. If the Bioloid kit is more than around $200.00 or maybe $250.00 I would personally shy away from it unless it is REALLY SPECTACULAR. If Bioloid makes their parts available for use, like for “replacements” and such, then I would like to explore this further and see if I can mate the Bioloid brackets with SES brackets.
I’ve seen the “Beginner’s Kit” available for $350 from two different vendors; this seems to be the smallest kit, with four servos (plus the controller, sensor module, battery pack, and misc. brackets). It’s enough to make a 4-servo walker similar to the Brat Jr., which can walk around and turn neatly before it hits a wall (or respond to sound, or whatever).
This just might be cheap enough for me to talk my wife into getting it for me for Christmas… maybe. But probably not this year (things are tight).
However, you can also buy a single AX-12 servo plus two brackets for $45. That’d be enough for you to see how SES-compatible it is, and if it is, it’s a dang cheap way to add a 16 kg cm servo to your SES bot!
(Of course you’d need to work out some way to control it, as it doesn’t respond to normal PWM signals… but my understanding is that it’s a simple, documented TTL serial protocol, so shouldn’t be too hard.)
Oh, and I found some info about the programming environment, which I ignored before because I don’t care. The latest Robot magazine includes this in its review, and there are also screen shots at various online vendors, like this one:
Basically it looks like the motion editor is a visual thing (complete with an onscreen 3D rendering of each pose, which is kinda neat), and then you have some sort of hand-holding programming environment where you combine the motions with sensor conditions to make behaviors. People who’ve used it seem pleased with it, but then, they’re Windows users.
Well, that is definitely not spectacular enough for me to want to save to purchase that kit. I think I can do a lot more with that sort of money buying SES stuff and building from scratch. Sure, I’ll probably spend more money, but I can do it all “one piece at a time” which works for me.
I’ve been programming ever since Junior High School, and have worked in 8 different computer languages.
Santa would have to bring me a kit before I would want to get one. I’d happily take one as a gift though.
This might appeal to me at some point, but I already have plenty to keep me busy right now. Have you checked out my new “W.A.L.T.E.R.” thread in Rovers yet?? It looks like I will be pushing off my build of Walk 'N Roll a bit, because now I really want some more stuff for Walter.
It sounds like it could be i2c. It’s a serial protocol, so I wonder if Bioloid used something standard. I will definitely have to look into this further. I can think of some places it might be really nice to have that sort of power in a robot…
Yep. I subsribe to Robot, SERVO, and Nuts and Volts. I read the article, but I am going to have to take a closer look now.
I don’t really care much about the hand holding, except when I am just starting to learn a new system. I am more into the pure power and flexibility of a programming environment and it takes a LOT to make me do more than yawn at a new programming system.
I just started fiddling around with Python again, and I have the serial and parallel addons installed for Windows and have to get them for Linux and FreeBSD also. When I get a robot with some sort of wireless like Bluetooth, I can see processing data from it and sending data and commands back to it using Python. I really like Python.
I’ve been programming ever since I was in Junior High School, all through High School (I graduated in 1976), and ever since. I’ve worked in at least eight different programming languages.
Oh, and Sun Microsystems is going Open Source with JAVA! I am very excited about this!!
I hear you — I’m in the same boat. It is a pretty good value though, considering you get the controller, battery pack, charger, brackets, and a very nice sensor package along with those very strong servos. But you’re right, $350 is a pretty high starting price.
Yes — very neat. Keep up the good work!
Again I agree, but people in this thread were asking about it so I dug it up. From your handle, I’m guessing you don’t care about Windows software much more than I do (I’m a Mac user). I’m also a software engineer, so writing code is no big deal… though I probably wouldn’t take time to write a sequencer as fancy as this one.
But, here’s something that is pretty neat, when it comes to building motion sequences: according to this interview, you can pose the robot physically (i.e., just move the actual robot how you want it) and then record this as a pose in the motion editor. Now that would be really handy, if you’ll excuse the pun.
If I were going to buy a build-it kit, I think I’d rather go spend $250.00 on Lego Mindstorms NXT kit. I’d have to see something REALLY special about the Bioloid kit in order to want to save up that much money for it. It would take two months for me to afford to get this.
Thanks! I am fine tuning Walter’s turret scanning behaviors now and making sure it falls through to the lower level behaviors when appropriate.
Actually, I have no real preference for any software system. I care about the software that helps me do what I want to do and is most flexible about letting me do things the way I want to do them, regardless of which OS it runs under. I use what works best for me.
, you can pose the robot physically (i.e., just move the actual robot how you want it) and then record this as a pose in the motion editor. Now that would be really handy, if you’ll excuse the pun.
One of the Open Servo guys is doing the exact same thing using the feedback features of the Open Servo. It’s a great thing to be able to do, but I can get that funcionality by using Open Servos which I have full access to the hardware and software (including full souce code, schematics, board layouts, etc) for.
I’ve been playing around with technology ever since I was in High School, so I guess I am not real easy to impress when it comes to commercial technology type products.
That is good news . Several months ago before I purchased the Lynx Hex Robot and now the Brat, I was flip flopping on which way should I go. I like the Robotis Ability to make several different things with the kit (I wanted to make the puppy). I also liked the digital servos that you could set and query different values. I also thought the daisy chaining was nice as it made neater looking robots…
There are several reasons in the end I ended up going with the Lynxmotion systems.
Probably the biggest one was support. With both systems you have great forum support by other robot builders, but with the Lynxmotion robots, you have the Owner here helping us. He also appears to be having as much fun as the rest of us. With RobotIs (RoboSavvy forum) there was only something like ten posts from anyone from Robotis and the last one was maybe five months ago.
The next biggest reason was expansion. I believe that the CM-5 controller is a Atmel AVR128. They even provide you with some sample code that shows you how you could program this in C. However in this version of their controller (unlike their CM-2) they did not expose any of the I/O pins that would allow you to easily expand to non Robotis parts. For example there is no easy way for you to add a gyro, Sonar, Infrared detector or any I2C devices. I have been following the web posting of several people who are trying to overcome this. So far there have been two approaches to this. One is to build a replacement controller board that totally replaces the CM-5. The other is to build a device that fits into the Dynamixel chain and add whatever expansion capabilities that want into this device. It seems a shame that Robotis has not come out with a version of the CM-5 controller that has these expansion capabilities.
My last reason was there was no North America distributor for it. But that sounds like that may change soon !
So who knows maybe someday I may have to revisit purchasing the Bioloid!
But then you couldn’t build anything with more than three servos. That’s a pretty severe limitation, especially if you want to make walkers. If LEGO had made serially addressable motors to remove that silly limitation, I’d be jumping on NXT in a flash.
Sure, but can you show me an Open Servo with 16 kg cm of torque for $45? I mean this as a serious question. If so (or even close), then that’s a serious contender.
So have most of us. No need to keep whipping out the ruler to try to impress us.
That’s the main reason I didn’t get the NXT to start with for doing robotics. It is too limited for my liking. There is the VEX stuff also, which looks interesting. Right now though, I think I can do pretty much anything I want to do with the SES stuff, and I am just starting out with that. I will probably have to have some custom SES brackets and such made for some things I want to build, but now that I am getting better with 3D CAD, I think I will be able to handle that when the time comes.
The nice thing about the Open Servos, is it can be applied to any standard servo. So, I am not limited to just one model of Open Servo. I also have access to all the hardware details (schematics, board layouts, etc) and software source code. Open Servos use Atmel micros. I doubt I could touch that spec with an Open Servo for $45.00. Are you talking about a standard sized servo or one of the Bioloid or similar ones?
One of the Open Servo guys is going to build mine for me when I am ready, including full assembly and modification of the servo(s) for $25.00 per servo. I just have to buy the servos I want to use and send them and money to him for each servo.
I am following this closely. Yes the Bioloid system has some cool features. I have decided to bring it to our store, and have already started the procedure. The big difference between the Vex, Bioloid and Lego NXT systems, and our SES system is we are continuously expanding the SES! I have 6 SES compatible GHM and PGHM motor mounts planned. We are designing an SES compatible linear actuator. Micro and quarter scale servo brackets are in the works too! When Hitec eventually releases the 475 and 645 “robot” servos, we will have the same features, just waiting for them…
Anyway the aluminum SES system is always going to be a more professional construction platform compared to Lego and Bioloid plastic, and the heavy Vex steel parts. Besides you can’t make this…
Well, when you going to stop planning them and make them for us? I have been thinking about venturing into the world of motors for some of my robot projects, but I need SES compatible mounts… Are you open to customer submitted designs for SES compatible parts?
I stop visitng the forum regularly for two days and I see this bombshell:
Holy!
I for one would really like to see those micro or mega brackets. I honestly don’t know which I would like to see first. I want micro ones for my tiny hexapod project (how about a micro version of the wicked leg?) and mega for a larger biped project. Of course, I also wish I had the time to finish my current projects, let alone start those. I’m sure many people here could use the motor mounts.
As far as the Biloid discussion, my 2 cents is that the only good thing about the kit is the servos are fast (unless that video was speed up) since the bot seemed to be walking much faster than a Robonova, the SES videos on the site, or I have been able to get my Scout to go with 5955s. Again maybe it only seemed faster. If carrying the kit is easy/ gets more people to come here, (or is SES compatible so people might buy it then upgrade) then by all means carry it. However, it doesn’t seem like a lot of people are buying the Robonova (from forum posts) and I can see work on supporting the Biloid slowing down development of the SES.
My preliminary plans are to create the equivalent of the ASB-04 and ASB-09 made from 0.09" aluminum for the quarter scale servos. The same hole pattern on them to allow connecting to all the other SES stuff. The first illustrated use would probably be the shoulder on an arm.
The micro servo supported will be the HS-85. I am planning ASB-04 and L brackets instead of the ASB-09 parts for pan and tilts. This will be much lighter on projects where weight is a major concern. Of course the entire SES set started as a pan and tilt, so we will see where the micro servo stuff ends up.
The KHR-1, Robonova, Bioloid Biped, etc all have followed the minimum weight and size scenario. It costs less tom manufacture and appears to work well. However adding additional axis is difficult to impossible to implement and or make functional. They all unfortunately appear to have closed the doors where ever possible concerning electronics expansion or alternate programming. It looks as if they are going the Compaq PC route of the past where they tried to force users to buy their peripheral electronics and all of their users suffered. Proprietary hardware and software only hurts the development of the technology.
Hm… with that information I’d say go with the mega first. Micro with the same hole pattern and horn size as standard is going to cause problems since they can’t really be made as small as they should be. Mega with the same hole pattern and horn size might seem a little bizarre (depending on the implementation), but for all intensive purposes shouldn’t be an issue.
I’ve looked briefly for mega servos which were more powerful than high end standard size, and actually haven’t really found any. I did find a few hobby servos which were slightly less powerful than a 5955. Then I found some industrial servos which were roughly the same size and quite powerful, but not really the standard mega form factor. Besides making it slightly cheaper to build an arm, in addition to the brackets we’d also need some monster servos to go with them. (for most projects)
Mega versions ASB-04 and ASB-09 are great, might I suggest trying to get mega versions of ASB-15 and ASB-17 soon after. As you might see, my goal is to basically make a Scout out of mega servos. The goal would not be for more torque, though higher torque/ faster servos would be nice, but to be able to fit more advanced electronics on the bot.
I’m going to end up writing my own sequencing and control software pretty much no matter what hardware I use. But I realize that others may not be in the same boat.
I like that about you though.
