Xevelabs

I received recently some Dynamixel XL-320 servos for beta-tesing. These are the new mini servos of Robotis' Dynamixel family. They are intended to be used for small robots that do not need a lot of torque, like the cute Darwin-Mini. Yet, what I would really love to do with them is to put them on a bigger robot, with AX and MX servos altogether. I envision a bot where joints could have the proper size of actuator, all the way from the MX-106 for knees and load baring joints to the XL-320 for fingers, sensors, or cosmetic movements (let's make the ears of Darwin-OP wiggle!).

Interfacing with AX / MX servos

Yet, there are 4 major differences that makes these small actuators not so simple to use with their bigger brothers:

• they are intended to run at 7.4V instead if 11.1V, meaning you can't just put them on the same power supply rail,
• they have smaller, 2mm pitch Micro-Latch Molex connectors that are incompatible with the Molex Spox 2.50mm of the AX and MX,
• they use the newer Dynamixel Protocol v2.0, same as the much much bigger Dynamixel Pro. (This was not the case for some earlier beta testers, but since the support website page has been changed, I believe it's going to stay like that from now on).
• they do not have regular mounting holes for M2 screws, they are intended to fit with OLLO plastic parts and their plastic rivets, meaning they have 4mm holes every 6mm (where AX have 2mm holes every 8mm, and MX have a variety of (non-compatible >_<) hole patterns).

When it comes to the first and second problems, a solution can be found with an adapter board. Aaron Perkins did just that, and I intend to make one too at some point - a smaller and less expensive one though.

The third one is not that bad, since commands in both protocols can be sent in turn on the same line, and there is no risk of misunderstanding. Only difficulty is that the master has to be able to speak fluently in both protocols.

When it comes to interfacing the servos mechanically with other stuff, there are multiple ways to go at it. The most obvious would be to use OLLO parts and rivets, just as intended, but this is not compatible with AX/MX servos, and there is a lot of give in every assembly. Rivets are great for small fun robots or prototyping but n match for screws when you want the thing to stay like that and not wobble around.
In the Bioloid Premium kit, they use a few OLLO-compatible parts, for example in the head to fit the ZIG-110A radio module. This is achieved with what look like M2.5 thermoplastic screws. They have a head with a diameter of 4.3mm, which is enough not to go through the 4mm holes of the OLLO parts.
M2 screws have flat heads around 3.6mm of diameter, but with the help of a small M2 washer (diam 5mm), they can be used to bolt the OLLO parts to regular AX parts. 5mm happens to be exactly the size of the small recess in the parts, in which the shoulder of the rivet usually fits.

Still, the difference in hole pitch (one every 8mm on AX servos, one every 6mm on the XL-320) will necessitate more work to use them together. The smallest simple pattern to fit OLLO parts on AX parts would be to put 5 OLLO holes (4x6mm = 24mm between the two further apart) aligned with 4 AX holes (3x8mm = 24mm), but this arrangement is not very practical. Also, 3 Ollo holes align well enough with 2 AX holes in diagonal to attach (like in the picture), yet I can't see it as a real solution for most cases...

I guess I'll work on some 3D printed mechanical adapter parts.

Other thoughts

The gearbox is 100% plastic, even the motor gear. However, there is a nice feature on the output gear: a torque limiter (the sort of double lobe thingy inside the biggest gear in the photo). This should make breaking the gears less likely in case of shock (like having a robot fall from a table).

The build quality is really good, the architecture is no-fuss, with one PCB on which the motor, STM8 microcontroller and potentiometer are soldered directly.
The sensor is an old-school potentiometer, like the good-old AX servos. There is enough room to put a magnetic sensor like in the MX servos, but the price would have certainly been much higher - such sensors cost maybe $4 more than the potentiometer, and I'm not even counting the magnet and increase of assembly and testing cost.

Some more references

Here are the references of the connectors for the XL-320 (that can be found also here).

• Male header, 3 ways: 53253-0370 (vertical), 53254-0370 (right angle)
• Female housing, 3 ways : 51065-0300
• Contact (24-30AWG): 50212-8000 (reel), 50212-8100 (bag)

More details on the XL-320 page on Robotis support website

The USB2AX project is still on track (more on that in a next post). As a little side-project, I wanted to have a try at making the USB2AX even smaller while retaining all the current characteristics (ATmega32u2, ESD protection, LED, etc). That's how the USB2AX Mini came to be.

The motivation is just to have fun and satisfy my lust for optimization. Or some might say pseudo-optimization here since by over-optimizing size, some other aspects of the device suffer, like quality of the physical connection (resistance to vibrations and yanking). But hey, this one will most probably never go beyond the prototyping stage, so please leave it to that.

I wanted to stay away from some obvious ways to make it smaller, like putting components on both side, using components too-small-to-be-soldered-with-my-crappy-prototyping-tools, requiring custom mechanical parts (like the ones used in the miniature bluetooth or WiFi dongles, with the electronics inside the USB plug) or cutting functionalities (ESD protection, status LED...). Thus it had to be possible to assemble with the same tools I already use and keep the nearly the same BOM as the v3.0. Working on the connectors seemed like the way to go, as they occupy around 75% of the board real estate.

The ideas here are:

• use a PCB USB connector, with gold fingers this time (saves 10mm)
• move the dynamixel connector so it is the closest possible to the USB plug, on the other side of the electronic components (saves around 5mm)
• route the USB lines between the Dynamixel connector's pads (saves still a little bit more)

And that's the result:

With a total length of around 22mm (compared to the 37mm of the V3.0), it already looks a lot smaller. However, by moving the Dynamixel connector a few millimeters inside the PCB, the actual length when measuring the device with a Dynamixel cable plugged in is 17mm shorter (41mm for the v3.0)!

From an electronics standpoint, it's nearly identical to the USB2AX v3.0. The only thing that changed is that instead of protecting each line of the serial port with a 47ohm and the anti-ESD chip, and then connecting them, I connected them directly then protected the remaining DATA line.

The PCB USB plug has, as expected, some disadvantages. Without modification, you have poor or non-existent electrical contact, that can be fixed by increasing the plug thickness with a piece of tape (here, I use 0.5mm thick Teflon tape). I could have had the PCB made with 2mm thick material, but it costs a lot more with Seeedstudio Fusion. The problems I used to have (see v1.0) with the electrical contact quality have been fixed by getting the board gold-plated (ENIG), with additional gold fingers on the contact pads.

It feels like you are plugging the Dynamixel cable directly into the USB port :P

I tried to protect the board with some heat-shrink tubing, but it tends to slip when shrinking, I had to do it 3 times to get this result.

All in all, I think it's a fairly cool little board, and it still does its job, so I'm happy with it :) As always, code and files are available in the Git repository :)

A lot happened since last post, and things are getting very close to their final shape.

The robot got some brand new 3D printed motor mount (thanks i.materialise!), all the required electronics in the legs, a second camera and a rotating laser distance sensor (the Neato LDS, from the XV-11 vacuum cleaner robot)!

New 3D printed parts

From front to back row: a part used to lock the ball bearing in place, the lower bearing mount and the motor mount,

That's what the motor + motor mount + pulley looks like when everything is assembled. This goes inside the aluminum tube. The motor is taken between two layers of 0.25mm aluminum sheet. With this, there is no risk of deformation of the polyamide part, plus it allows for two M2 grub screws (both in the bottom of the part, one close to the front, one close to the back) to secure the motor and reductor in place without blocking the gears.

The motor mount and the part with the ball bearing are now two distinct elements. With the new motor mount design, it was necessary to split the two, as one is fixed and the other has to be adjusted to have the right belt tension.

My first thoughts when I designed the mechanism was that little to no effort would apply to move the bearing out of its hole, but torsion in the tibia support part quickly showed that it was not the case. This little part keeps the bearing where it is. Here it's demonstrated with an old motor mount.

Leg Control Board

Each leg is fitted with a Baby Orangutan B-328 that controls the motor, the magnetic encoder, the wheel orientation servo, and talks on the Dynamixel Bus as a real Dynamixel device. Right now, it masquerades as an AX-12+ and can communicate with Roboplus :)

In order to have it talking on the Dynamixel bus, I had to apply two important modifications on the B-328 board: change the oscillator to a 16MHz one, and remove the user LED (the red one).

Having a 16MHz oscillator instead of a 20MHz one is needed to join the bus at 1Mbaud. I found the perfect replacement here (Murata CSTCE16M0V53-R0) .

The LED had to go because it's tied to the ATMega328p's UART TX, and caused the idle tension on the line to be as low as 1.8v (using only the inbuilt PD1 pull-up resitor) when it should have been 5v. This does not play nice with the original usb2dynamixel, which often saw 0x00 bytes that were not there. Unsoldering (or... destroying -_-") the led solved the problem.

CMUcam3

Having a second camera might seem overkill, yet it addresses one of the most important sensory question for this year's competition: "what color is the square In front of me?". It also be used for line-following, and identification of the team's color at startup (one less switch to flip myself... yeah I'm lazy like that ;) ).

I will write a simple custom firmware for it, that will probably do segmentation (it's quite easy in HSV with the colors used on the playing field and playing elements), basic line following, and maybe some rough position tracking based on the grid. The observable area varies with body height and rotations, and the lens distortion is quite important in the corners. Fun times ahead to extract meaningful information ^^.

Owing to last year's disappointment with the light conditions during the approval phases (very different light conditions in different parts of the playing field, hard shadows, yellow lights...), I decided to add some powerful white LEDs on the camera to try to have at least a spot of well-lighted ground...

Neato LDS

The lidar is mounted upside down, below the body. During normal operation, the sensing plane of the LDS will be around 8 or 9cm above the ground, and it's aim will be to detect obstacles I can't go over (like the opponent :p ). I'm still working on a protective cage around the sensor to avoid any damage to it. I don't want to know if it can support the full robot weight ;).

As always, a lot more pictures are available in the gallery!

At some point last month, before I received any of the manufactured parts, I became tired of building the robot only as a CAD model, and wanted to make some physically tangible progress.

So I disassembled completely the Bioloid humanoid, and started assembling the Dynamixel part of the legs. The two plastic plates in the middle of the leg proved to be too flexible, so I replaced them with a 3mm aluminum part. I also took this opportunity to correct the alignment of the third servo axis, so that it intersects with the axes of the first two servo in a single point.

But that was not enough for my thirst of physical manifestation, so I built a prototype of the body, using black and white 2mm styrene sheets. It works as a prototype, but I would not even try to have it walk with such a flexible body.

I later replaced the bottom plate by a more rigid 3mm medium one.

Wiring still needs a little work, but I like where it's going :)

Having the things in my hands made me see how big it will be, and how I could improve a few things. For example, I adjusted the distance between the front legs so that the Beagleboard fits snugly with my shortened RS-232 connector plugged in.

I also added the CMUcam3 as a giant color detector, watching the ground. It will be in charge of looking at the grid to provide absolute positioning information every time the robot crosses a line, and also will be used to dock nicely on the playing element the robot wants to move. The idea is to have an easy alternative if I can't get a reliable positioning system using only the Neato LDS lidar, as it won't be parallel to the ground most of the time! Communication between this camera and the Beagleboard will use the shortened RS-232 showed above.

Tadaaa!

At first, the Xachikoma project was completely separate from my participation to Eurobot. However, building two complex robots at the same time would have been way too time consuming, and nearly impossible to financially overcome. That's why, as I hinted in previous posts, I chose to merge the two projects into an Eurobot-playing version of the Xachikoma.

Of course, the two projects had different goals: one needed to be able to play by this year's Eurobot rules, whatever its means of achieving it, when the other had to look and act accurately like a Tachikoma... Not exactly a perfect match!

So the idea is this: build a robot which has the possibilities of moving like a Tachikoma (legs with enough degrees of freedom (DOF) and powered wheels), fitted with lots of sensors and equipments to be able to score some points and not be ridiculously unadapted for the competition, all of this without giving too much importance to the cosmetic aspect. This way, I can work on the mechanics, electronics and leg movements this year, and make an accurate-looking Tachikoma later. Also, I don't want to show up at the competition with a robot so unadapted it would have to rely on luck to be accepted past the approval phase. That would be kind of wasting the time of the referees and other participants.

The main structural differences between the original robot and the one I'll make this year are these: the Xachikoma won't have a pod, and probably no arms or eyes either :/ . The body, while still round, will be significantly different from the cute rounded blue shell we are accustomed to. On top of the body will probably be a moving structure destined to maintain the beacon support at exactly the right height and inclination whatever the body movements. On the bottom part of the body, the Neato Piccolo LDS (Laser Distance Sensor, aka the lidar I referred to in previous posts) will be mounted head down, but this still needs some consideration...

The chassis of the robot is heavily based on the Bioloid Premium Kit I talked about earlier. This is the core of the robot: 16 AX-12+ (at least 4 of which will be replaced by AX-18F), plus custom parts for the body itself and the tibia. With the AX-12+ and even more powerful AX-18F, the robot will be able to easily bear the payload (Beagleboard-xM, Neato LDS, PSEye camera, controllers, batteries, dynamically stabilized beacon support, other sensors...).

For size comparison, here is the chassis (I just began the 3D model of the body) next to a tower of playing elements. The perimeter when standing like that is a little below the 120cm mark, and it can easily be shrunk or widened by changing the leg's position.

The leg has two parts: the first one is fully built from the Bioloid Premium Kit elements, and goes from the body to the knee. It is composed of 4 Dynamixel servos, arranged to have kinematics identical to the Xachikoma prototype, and to the Tachikoma. Particularly, the axes of the first three servos are concurrent, and the axes of the last two as well.

The second part of the leg is custom made, and goes from the knee to the wheel. It's composed of (from top to bottom): a servo support (weird shaped white piece, 3D printed), a metal-geared digital micro-servo, a second servo support (second white part, 3D printed), the tibia cut out of a 20x20 aluminum tube, a motor with its bracket (yellow part), two pulleys(also 3D printed) and their timing belt (not shown), and a wheel. Add to this a force sensor to know when the leg is on the ground, a magnetic encoder for motor control, probably one or more contact or proximity sensors, and a Baby Orangutan to manage all of this.

The wheels are nice and have good traction. Only problem could be the color (green!) ^^. The rotation axis of the small servo passes through the contact point of the wheel with the ground, which has the advantage of not affecting too much the kinematics of the rest of the leg.

The servo supports are tricky parts: they must allow for +/- 90° rotation of the lower part of the leg, and yet be able to support all the weight of the robot, even with occasional shock. And on top of that, as it's the outermost part of the leg, it has to be as small as possible to avoid breaking the perimeter limit. It also mustn't get in the way of the tibia movement by reducing the range of the other servos of the leg. All these consideration led to this (arguably efficient) design.

They will be printed in ABS plastic by the cool guys at i.materialise. ABS is a strong and rigid material, with a is very dimensionally stable and accurate printing process (meaning the part has little risk of being deformed while printed). Using Finite Element Analysis, the profile of the part has been refined to remove as much as possible the weak points of the design. Here is what I got on one of these tests: As you can see, the stress is spread across the whole lower part, with two potential small weak spots (in red). The first iterations had just one big crimson area at the bottom of the part, so this is an improvement ^^. Have a look here for an interesting tutorial about FEA.

The pulleys are more of a bet. I wanted a way to drive the wheel without slipping, yet I did not have enough space to fit gears, or commercially available timing belt pulleys... so I will try to print them with white polyamide (called "white, strong and flexible" by Shapeways).

The belt will be a 225 5M HTD (225mm length, 5mm between two teeth, High Torque Drive), but with non conventional width. I will try to reduce it to around 3mm, instead of 9. This would allow for pulleys to be way smaller (7mm width instead of the commercially available 20mm). I redesigned the pulleys from what I could find about the 5M HTD specification (ie. not much), and took this chance to make them just the way I need to mount on my mechanism. For example, I added a small ring on the side of the pulley that will allow perfect centering of the multi-pole magnetic ring that goes with the magnetic encoder (the blue ring). The pulley on the wheel will be mounted with six nylon screws onto the wheel, while the one going on the motor will fit on its shaft with a M3 grub screw.

Quite a few details are still in the design phase (like the body... a big detail!) but this is all I can show right now :)

In the next few weeks I should finalize this design, have the plastic parts printed, the aluminum part CNCed and the other parts shipped or hand-made. To be continued !