I just wanted ot make sure i have everything that I will need to build the servo erector set version of the EH3-R. Comments and Suggestions/recomendations are welcome and enouraged.
1x programming guide for the BS2-E that includes BS2-E and Bot board with detailed instructions.
1x SSC-32 with SEQ visual sequencer for ease of programming(no programming expereinece)
6x 3DOF legs
2x 6v 2800 battery packs
1x Lexan black EH3-R main chassis (will it work with 3DOF legs?)
18x HS475HB servos
Any suggestions for sensors to make this robot autonomous would be greatly appreciated. I want it to be able to live out a battery pack and maybe later try to get somewhere everytime you turn it on. I don’t know. just basically I would like an autonomous robot that trys to explore its environment,l and also, do they make infared camera sensors and humidity sensors? Thanx so much guys.
You may want to seriously consider using the Atom instead of the BS2-E. There are programs already written for the robot using the Atom. The program is extremely powerful, and represents countless hours of work. The inverse kinematics alone will save you dozens of hours. I would hate to see you have to reinvent the wheel, and frankly the BS2-E is under powered for this robot. The existing program for the Atom is for remote control using a play station wireless game controller. But these RC routines can be removed and a sensor routine can be put in it’s place.
I also strongly recommend you use all 475 servos if at all possible. These are a minimum for this robot. I wouldn’t run it with two 2800mAh batteries. Save some weight and use 9vdc or 7.2vdc 1600mAh batteries for the logic. The chassis is available in black and of course it works with the 3DOF legs…
Than alot guys! The reason I wanted to get the BS2-E was that the Lynxmotion sequencer can generate code for it. So the Atom has inverse kinematics built into it? prices? If the SEQ could also generate ATOM code then I don’t have any reason left not to switch…except for maybe prices.
yep just checked into it, as long as it can be written with the SEQ or another visual representation “Big blocks” software I am fine with it. Programming scares me…lots
Is the kit for the 3DOF legs more expensive than buying all parts individually? I don’t remember having to spend ninety bucks for a pair of legs when I factored up the cost from the legs and chassis. not accusing you guys of price gouging or anything like that I think your prices are very reasonable, just wondering.
I sthere anything available like the SEQ for the ATOM that I could use to simplify programming? If I could use inverse kinematics maybe then I could do it? hmm… Programming scares me …lots
Trust us.
Just about anyone can pick a microcontroller and learn it’s programming language.
Sure… you’ll run into frustrating problems and pull out a bit of hair.
But, if you can learn how to play a video game, you can learn how to program.
Both are learned best the same way: jump in, die countless times, get frustrated, throw the controller, go online, get cheat codes (no pun intended), and then go back to getting your butt kicked (but this time, with infinite health, so it doesn’t matter).
Eventually, you master the game (or the programming, in this case) simply by wanting to do it enough.
And there’s good news, which you seem to have misenterpretted a bit.
Lynxmotion provides free cheat codes!
You asked whether those “cheat codes” come built within the Basic Atom.
No, actually, they don’t.
You go to the main website, click on information, locate your hexapod’s tutorial, and find the example code for it there.
Then you download it and load it into the Basic Atom when you get it.
Besides the huge amount of example code on the Lynxmotion website and the Basic Atom’s website, you also have us here!
Don’t feel bad about asking really dumb questions (I’m sure I asked them, at some point).
At worst, we’ll at least point you to where you can find the answer.
Does anybody know of any long-range infa-red sensors for robots? I want to make a hexapod that chases/scares the cat and dog via following them and making scary noises.
so what you are sort of looking for is a PIR array with a frensel lens that breaks the field of view into angular sections, with which you could then overlay that data with ultrasound information to find the moving “body temperature” items in a room?
actually, if you set up a few PIR channels and swept it across an angular range that overlapped the angular field of view for each sensor you could extrapolate the position of the warm bodies without their even having to move, and this would be cheaper to implement than the previous paragraphs idea/concept. it would look like some kind of targeting radar head too.
Well, both the IR and the Sonar rangefinders get the same sort of information: how far the sensor is from the object in front of it.
The differences between the two types arise from how they find this distance.
The former uses Infa Red, and the latter uses sound.
Both objects “ping” a signal out and find how much time it takes for the “ping” to come back.
So, to answer your question, neither an IR nor a Sonar is the best option for you.
You’ll probably want to talk to Mike (SN96) about using a wireless camera and free imaging software (can’t remember the site off the top of my head) to track objects (a cat, in your case).
So this camera would pan around and once it found something I designated then it would lock onto it and tell the microcontroller to do it’s thing? could a basic ATOM handle that kind of task? seems a bit much for such a little chip…
With a camera operating in the visual spectrum you are going to have a hard time distinguishing between a gray cat and a pile of dark laundry (socks) on the floor unless you have a LOT of processing power. You would likely want a wireless camera transmitting back to a host workstation where the image processing would be attempted and command instructions would then be sent back to the 'bot. By shifting out of the visual spectrum many things look different enough that target identification and acquisition tasks becomes much more managable, although I would still be surprised if you could do it on anything less than a workstation or dedicated DSP platform. There are CCD cameras though that with the appropriate IR filters can certainly operate in the near infrared, or if you have money to burn actual long wave IR (1100nm+).
could I just send it back to my pc via a bluetooth and have my labtop take care of the visual spectrum distinction stuff, then like tell the microcontroller if it was or wasmn’t a cat and the microcontroller would make it’s next desicion based on the positive/negative signal form the laptop PC?
Programming scares me…lots 
Programming scares me …lots 
