Monthly Archives: December 2014

trashbot 3: a gimbal-like autonomous 2DOF spine

(if you can’t wait to see the end result, scroll down to the bottom watch the second last video šŸ˜‰ )

after the lower body was walking and free from seizures when starting up the arduino, it was time to think about the rest of the robot. certainly, there’s still much to improve on the gait and most importantly the feet, but it’s my goal to build up the complete robot first before iterating the parts (the feet are under constant change as the weight distribution varies and also i’m constantly learning how to improve gait pattern.)

in contrast to many commercially available robots (like the bioloidĀ or the lynxmotion pete), i wanted to have a machine that feels more natural and can move a bit more freely, especially with the head and the body trunk. usually, these two are rather stiff or absent (head). people invest dozens of servos into legs but not a single oneĀ into the spine. for me, the two servo legs are “good enough” forĀ the moment, i’ll invest in the rest first.

on the other hand i’ve seen many gimbal mechanics mostly for quadcopters carrying the camera. some of them were even arduino controlled and here and there, you can see some code. interestingly, i wasn’t really able to find servo-driven automatic gimbals, most of those are driven by dc motors.

let’s view the goal first:

okay, here’s how i built it:

the base was taken from anold bike light:

bike light

and i added the first servo that can swing from left to right:

lower servo for the back

next up, was the second servo driving the chest allowing the bot to lean forward. it seemed to me that these are the two directions the biped needs to be stabilised in, since the feed generate a tilt to the sides and the hips rotate body on that tilted axis.

i wanted to have a possibility to update the mechanism easily and didn’t want to glue or solder anything, also bulky servo brackets seemed not the right way to go aesthetically. so i came up with a wire-based connection of the two made from a recycled wire that i had left from hanging lamps and some power connectors and a spring from the recycled typewriter:

connecting chest and back servo

it took a while to find the right parts from the type writer and re-arrange them so that i had a broad enough shoulder that could work as a “servo bracket” while being stable enough to support more hardware on top later on. Ā and i recycled my first two letters “M” and “N”:

chest and should bracket

finally, connecting the two parts to this:

upper body assembled

it was really tricky to find some parts to surround the upper servo so that i could atttach a second ankle on the other side of the servo. kind of proud to have it included in the wire fixation structure… also, you see a recycled air pump holder from an old bike that protects the bot’s chest servo should it fall backwards.

next i recycled a heat pipe from an old (TV?) pcb:Ā heat pipe

and shaped it to support the upper body on top of the hips and TADAA! here’s the assembled trashbot 3:Ā trashbot v3 mechanics front

trashbot v3 assembled mechanics side

now it was time to develop the electronics and the software. (actually, i developed the software with the spine detached and assembled the whole beast when the autonomous gimbal was actually working).

standard setup with the gyro / accel MPU6050, most importantly with the interrupt attached to the arduino:



i tried a could of libraries for getting decent values (including the kalman filtering that can be done on the DSP of the MPU itself), but jeff rowberg’s is simply the best.

finally i wrote my own PID functions, mostly to understand how the maths work and how and whether i can fine tune something. PIDs are beasts as they are designed to self balance to a set value. now if the actuator and the sensor are attached to each other and controlled via a PID the whole system can swing like HELL. actually i burnt my first servo playing with the setup and was happy to learn that my humble mechanic design worked really well for maintainability…

i used mostly these links to understand how it works:

and lots of reading around this. the problem is that nobody can tell you how to tuneĀ your PIDĀ inĀ your setup. so there’s a whole lot of trial and error with many strategies being published on how to find the right parameters. as you can see in the video, the parameters are sort of okay, but still the system is behaving a bit “choppy”. but if i react too often to the sensor values, the system starts to swing. so here’s some homework left and i hope that this blog post will spur some discussion in the forums. and there’s a nice blog post that obviously led to the PID library included in the arduino software distribution. i also want to learn from that.

let’s see the system in action:

finally, i also studied a bit how the bot walks with and without the gimbal active. the first part of the video shows it from front and top with the gimbal active, the second part shows it from top and side with a stiff upper body.

you can see that the upper body’s lateral amplitude is higher when switched off. this will be even more the case when we load the shoulders with a head and arms in the next iterations.

i will do another blog post investigating the gimbal effect more deeply. i just bough a tiny camera that i can attach to the neck and will tape trashbot from “first person perspective” šŸ™‚



trashbot version 2

it took a while to improve trashbot v1 to v2 (for the video, scroll down) as i had to do a couple of things first:

finally, i got decided to increase the trash share of the bot and get rid of the shiny new feet made from polystyrene and replace them with old PCBs. i also increased the length of the feet for better stability.

IMAG0859then cut out the feet:


also added the typewriter’s “shift” mechanics:


to the tighs as their angles were the major source of instability and shakiness:


i also improved the electronics after my recent experiments with relays:


from the top it looks like i just added the switch and the relay, but i also added more “ports” for four more servos (intended for the upper body) as well as a “port” next to the switch for an UBEC. see the real wiring here:


as you can see the servos (left up) get their power via the “port” for the UBEC (bottom) and then via the switched part of the relay. the relay is powered directly by the arduino (red cable coming from top, arduino port d13).

the trick is that the relay is cuting off the servo power when not switched. the arduino then has to set d13 HIGH to power the relay’s coil and thus keep the power on. since the d13 can only be powered via software, the arduino can control the servos when it has booted up and during the boot up the servos don’t behave badly. (obviously the d13 could also be in a HIGH state when the arduino powers up but that seems to be just to short for the relay to react.)

alright, after all the construction mumbo jumbo, here’s the end result:

compare that to version 1:

i also used a new servo library VarSpeedServo which can move the servos at user defined speeds. not sure i’ll keep it because i want to experiment with more complext gait patterns but for now it’s a great solution.



preventing a robot from seizures when powering up

during the work on trashbot v1 and rofi, i noticed that robots get seizures when powered up (see the second video in this post for examples). this is obviously because the arduino’s outputs are not in a defined state for a brief time and the servos also get power already leading to outputs sometimes firing until the arduino has loaded its bootloader and executes the user’s code.

some discussions on the arduino forumĀ and on google+ revealed that it might be helpful to add a relay.

so i ordered some JQX-118F relays from www.pollin.deĀ that cost 1.95ā‚¬ and switch by 5 volts.

here’s my first experiment, simply switching switching one led off and the other on. relays usually switch from one contact to another, meaning that even if the relay coil isn’t powered, there is a default state for both switches (one is on, one is off):

next i wanted to document how actually the servo moves into undefined positions when powered up, so that we see that this is a real problem to solve (yeah, i know that the 2s / 7.4v lipo may be too strong for the mg996r servo, will add a ubec soon):

okay, so finally, we add the two together and use one arduino output to control the relay coil, the relay coil then will switch the power to the servos. i will to this for the whole power rail feeding all the servos of the robot so that i only need one relay and one digital out of the arduino nano. let’s see how it works:

i will add the relay now to trashbot v2’s circuit and see whether there will be any more problems afterwards. i’m not sure whether i need something more (apart from the missing ubec to reduce the voltage for the servos) like diodes (coil loosing power gives it back to the arduino? or not strong enough?) or pull up/down resistors (have to teach myself what that’s all about next). but for now, i’m happy to see it working.


i just had to slaughter one of my typewriters to get more parts for my robots

afterĀ building the first of trashbot that contained mostly parts that i had floating around, i learned that i want to build more robots entirely made from residuals of other machines.

i found some more parts in my cellar, but actually i realised that one beautiful assembly is a mechanical typewriter from the sixties: the olivetti lettera 32 that i truely adore because of its compact design, the missing floor plate and of course its color.

i actually own three of these, so i thought that sacrificing one of them to live on in various robots is actually a nod to them by bringing them back to life. albeit, a different life.

here you see the german model (i have one spanish and one french model, too), the color is not original anymore as the typewriter seems to have lived a life in the sun…


getting the frame off was not easy as there were some hidden screws:


in the front / lower picture you see some of the springs the bring the key back up when pressed:


and the screws showed some resistance…


getting to the “core assets”: the letters and the keys:


actually hard to dig deeper into it without bending the metalhousing for the keys’ fingers:


but in the end i made it:


one of my learnings was that the keys have different lengths of their arms connecting to the lettersas you have to translate the square formation of the keys to the half-circle of the letters so that they can hammer the tape at the same position:IMAG0844

here’s the letters with the complex “SHIFT” mechanics on the left and right. bastards really to me a while to rip them off.



here they are, looking like creepy little bugs. turns out that they are the first parts to go into trashbot v2:


here’s the clean letter choir (beautiful, isn’t it?):


here’s the paper transport and roll disassemble, in the middle you see the curled spring for transporting back automatically when hitting “return”. in the back / top left you see the the bell for the “ding”:


finally, that’s the mechanic yield!


glad i finally blogged about this. in fact, trashbot v2 is alive already and will follow this weekend, stay tuned!