Friday, November 28, 2014

FingerTech Mecanum meets Actobotics

Sparkfun sell some nice little omniwheels made by FingerTech Robotics. These come with a grub screw mount for a 3mm axle. While it is said around the webs that one might drill out the mount to accept up to a 6mm axle, I wanted a more flexible solution for mounting these wheels onto an Actobotics structure. It turns out that the four screw mounts (using what I think are x-40 screws) are in an extremely close location to the four screws on the Actobotics hub mount. Unfortunately it was a tad hard for me to get a hold of longer x-40 screws to attach the hub mount, so I ended up taking the wheel apart and replacing the standoffs with the Actobotics ones. The result is shown below:


The below shows the mecanum wheel taken apart. The three standoffs you see vertical in the image are the original ones from the wheel. These are about 1 inch long, so you'll be wanting some 1 inch actobotics standoffs to replace them with. When you unscrew the original standoffs then the hub mount part (centre of the red alloy), will be able to fall out and be removed. This lets you screw the Actobotics standoffs on and then on the other side use slightly longer bolts to attach the hub mount to the wheel to get the assembly shown above.


Apart from a clean 6mm hookup for the stepper motors that I plan to use, this is a handy modification allowing the 1/4 inch hub mount or other sizes to be substituted in instead. This is handy if you want to switch from 6mm to 6.35mm (1/4 inch) axles as you can easily change your mind just by changing the actobotics hub mounts.

Sunday, November 23, 2014

Terry: Updated Top Shelf

The Kinect is now connected much closer to the tilt axis, giving a much better torque to hold ratio from the servo gearbox. I used some self tapping screws to attach the channel to the bottom of the Kinect. Probably not the cleanest solution but it appears to mount solidly and then you get to bolt that channel to the rest of the assembly. For a closer look the Logitech 1080 webcam is mounted offset from the Kinect. This should give an enjoyable time using the 1080 RGB data and combining the VGA depth mask from the Kinect into a point cloud.


The camera pan/tilt is now at the front of the top shelf and a robot arm is mounted at the back of the shelf. The temptation is high to move the arm onto a platform that is mounted using threaded rod to the back of Terry. All sorts of fun and games to be had with automated "pick up" and move tasks! Also handy for some folks who no longer enjoy having to pick items up from the ground. The camera pan/tilt can rotate around to see first hand what the arm is doing, so to speak.


The wheel assembly is one area that I'm fairly happy with. The yumo rotary encoder runs 1024 P/R and it is attached using an 8:1 down ratio to give an effective "ideal world" 13 bit precision. Yes, there are HAL effect ICs that give better precision, though they don't look as cool ;) The shaft of the motor is the axle for the wheel. It is handy that the shaft is not right in the centre of the motor because you can rotate the motor to move the wheel through an arc, and thus adjust the large alloy gear until it nicely mates with the brass gear on the rotary encoder.



Lower down near the wheels is a second distance sensor which is good for up to around 80cm distance. The scan rate is much slower than the Kinect however.


Things are getting very interesting now. A BeagleBone Black, many Atmel 328s on board, and an Intel j1900 motherboard to run the SLAM software.

Friday, October 31, 2014

Terry 2.0 includes ROS!

What started as a little tinker around the edges has resulted in many parts of Terry being updated. The Intel j1900 motherboard is now mounted in the middle of the largest square structure, and SSD is mounted (the OCZ black drive at the bottom), yet another battery is mounted which is a large external laptop supply, the Kinect is now mounted on the pan and tilt mechanism along with the 1080p webcam that was previously there. The BeagleBone Black is moved to its own piece of channel and a breadboard is sunk into the main 2nd top level channel.


I haven't cabled up the j1900 yet. On the SSD is Ubuntu and ROS including a working DSLAM (strangely some fun and games getting that to compile and then to not segv right away).

I used 3 Actobotics Beams, one IIRC is a 7.7 incher and two shorter ones. The long beam actually lines on for the right side of the motherboard that you see in the image. The beam is attached with nylon bolts and has a 6.6mm standoff between the motherboard and the beam to avoid any undesired electrical shorts. With the two shorter beams on the left side of the motherboard it is rather securely attached to Terry now. The little channel you see on the right side up a little from the bottom is there for the 7.7 inch beam to attach to (behind the motherboard) and there is a shorter beam on this side to secure the floating end of the channel to the base channel.



The alloy structure at the top of the pan and tilt now has a Kinect attached. I used a wall mount plastic adaptor which with great luck and convenience the nut traps lined up to the actobotics holes. I have offset the channel like you see so that the center of gravity is closer to directly above the pan and tilt. Perhaps I will have to add some springs to help the tilt servo when it moves the Kinect back too far from the centre point. I am also considering a counter balance weight below the tilt which would also work to try to stabilize the Kinect at the position shown.



I was originally planning to put some gripper on the front of Terry. But now I'm thinking about using the relatively clean back channel to attach a threaded rod and stepper motor so that the gripper can have access to the ground and also table top. Obviously the cameras would have to rotate 180 degrees to be able to see what the gripper was up to. Also for floor pickups the tilt might have to be able to handle a reasonable downward "look" without being too hard on the servo.

There were also some other tweaks. A 6 volt regulator is now inlined into a servo extension cable and the regulator is itself bolted to some of the channel. Nice cooling, and it means that the other end of that servo extension can take something like 7-15v and it will give the servo the 6v it wants. That is actually using the same battery pack as the drive wheels (8xAA).

One thing that might be handy for others who find this post, the BeagleBone Black Case from sparkfun attaches to Actobotics channel fairly easily. I used two cheesehead m3 nylocks and had to force them into the enclosure. The nylocks lined up to the Actobotics channel and so the attachment was very simple. You'll want a "3 big hole" or more bit of channel to attach the enclosure to. I attached it to a 3 bit holer and then attaced that channel to the top of Terry with a few threaded standoffs. Simplifies attach and remove should that ever be desired.

I know I need slip rings for the two USB cameras up top. And for the tilt servo as well. I can't use a USB hub up top because both the USB devices can fairly well saturate a USB 2.0 bus. I use the hardware encoded mjpeg from the webcam which helps bandwidth, but I'm going to give an entire USB 2.0 bus to the Kinect.

Wednesday, October 15, 2014

Sliding around... spinning around.

The wiring and electronics for the new omniwheel robot are coming together nicely. Having wired this up using 4 individual stepper controllers, one sees the value in commissioning a custom base board for the stepper drivers to plug into. I still have to connect an IMU to the beast, so precision strafing will (hopefully) be obtainable. The sparkfun mecanum video has the more traditional two wheels each side design, but does wobble a bit when strafing.


Apart from the current requirements the new robot is also really heavy, probably heavier than Terry. I'm still working out what battery to use to meet the high current needs of four reasonable steppers on the move.

Tuesday, September 9, 2014

Getting a feel for Metapolator and Cascading Parameter Values


Metapolator is a new project for working on font families. It allows you to generate many fonts in a family from a few "master" cases. For example, you could have a normal font, modify it to create a rather bold version and then use metapolator to generate 10 shades on the line between normal and bold.

I have recently been reading up on metapolator and how to hack on it. So this post describes my limited understanding of what is a fairly young project. So warnings inline with that are in place; errors are my own, the code is the final arbitrator etc.

Much of the documentation of Metapolator involves the word Meta, so I'm going to drop it off this post as seeing it all the time removes its value in terms of semantic add.

At the core of all of this polating are parameters. For example, after importing a font and calling it "normal" you might assign a value of 100 to xheight. I am assuming that many of the spline points in the glyph (skeleton) can then be defined in terms of this xheight. So the top of the 'n' might be 0.95*xheight.

A system using much the same syntax as Cascading Style Sheets is available to allow parameter values to be set or updated. Because its parameters, its called CPS instead of CSS. So you might select a glyph like 'glyph#n' and then set its xheight to be 105 instead.It seems these selectors go right down to the individual point if that's interesting.

In order to understand the CPS system I decided to start modifying a basic example and trying to get specific values back out of the CPS system. The description of this is mainly to see if my playing around was somewhat along the lines of the intended use of the CPS system.

For this I use a very basic CPS

$ cat /tmp/basic.cps
 
* {
     label : 1234;
     xx    : 5;
}

glyph#y penstroke:i(0) point:i(0) {
     xx    : 6;
}

$ metapolator dev-playground-cps /tmp/basic.cps

The existing dev-playground-cps command makes its own fonts up so all you need is a CPS file that you want to apply to those fonts. In my case I'm using two new properties, the label and 'xx' which are of type string and number respectively.

A default value of 3 is assigned to xx for all points and each point and glyph get a unique label during setup.

I found it insightful to test the below with and without selectors that modify the 'xx' property in the CPS, and at both levels. That is, changing the xx:5 and xx:6 in the above CPS to be xxno1:5 and xxno2:6 and seeing what the below printed out. The xx.value makes the most sense to me, show me the default value (3) if nothing is set in any CPS to override it or show me what the CPS has set if it did any override for the point.

element = controller.query('master#heidi glyph#y penstroke:i(0) point:i(0)')
console.log('element:', element.particulars);
console.log('element:', element.label);
console.log('element:', element.xx);
computed = controller.getComputedStyle(element)
console.log('label: ' + computed.get('label'));
console.log('xx.base   : ' + computed.getCPSValue('xx'));
console.log('xx.updated: ' + computed.getCPS('xx'));
console.log('xx.value  : ' + (computed.getCPS('xx') ? computed.getCPS('xx') : computed.getCPSValue('xx')));


The above code is also pushed to a branch of my mp fork at cps.js#L213

I found that a little tinkering in StyleDict.js was needed to get things to operate how I'd expected which is most likely because I'm using it wrong^tm.
The main thing was changing getCPSValue to test for a local entry for a parameter before using the global default StyleDict.js#L93.

I might look at adding a way to apply a CPS to a named font and showing the resulting font as pretty json. For reference this will likely have value and valuebase showing the possibly CPS updated value and the value from the original font respectively.

Thursday, August 28, 2014

Terry is getting In-Terry-gence.

I had hoped to use a quad core ARM machine running ROS to spruce up Terry the robot. Performing tasks like robotic pick and place, controlling Tiny Tim and autonomous "docking". Unfortunately I found that trying to use a Kinect from an ARM based Linux machine can make for some interesting times. So I thought I'd dig at the ultra low end Intel chipset "SBC". The below is a J1900 Atom machine which can have up to 8Gb of RAM and sports the features that one expects from a contemporary desktop machine, Gb net, USB3, SATA3, and even a PCI-e expansion.


A big draw to this is the "DC" version, which takes a normal laptop style power connector instead of the much larger ATX connectors. This makes it much simpler to hookup to a battery pack for mobile use. The board runs nicely from a laptop extension battery, even if the on button is a but funky looking. On the left is a nice battery pack which is running the whole PC.

An interesting feature of this motherboard is no LED at all. I had sort of gotten used to Intel boards having blinks and power LEDs and the like.
There should be enough CPU grunt to handle the Kinect and start looking at doing DSLAM and thus autonomous navigation.

Monday, July 14, 2014

Hookup wires can connect themselves...

A test of precision movement of the Lynxmotion AL5D robot arm, seeing if it could pluck a hookup wire from a whiteboard and insert it into an Arduino Uno. The result: yes it certainly can! To be able to go from Fritzing layout file to automatic real world jumper setup wires would have to be inserted in a specific ordering so that the gripper could overhang the unwired part of the header as it went along.


Lynxmotion AL5D moving a jumper to an Arduino. from Ben Martin on Vimeo.

Saturday, July 5, 2014

Ending up Small

It's easy to get swept up in trying to build a robot that has autonomy and the proximity, environment detection and inference, and feedback mechanisms that go along with that. There's something to be said about the fun of a direct drive robot with just power control and no feedback. So Tiny Tim was born! For reference, his wheels are 4 inches in diameter.


A Uno is used with an analog joystick shield to drive Tim. He is not intended and probably never will be able to move autonomously. Direct command only. Packets are sent over a wireless link from the controller (5v) to Tim (3v3). Onboard Tim is an 8Mhz/3v3 Pro Micro which I got back on Arduino day :) The motors are driven by a 1A Dual TB6612FNG Motor Driver which is operated very much like an L298 dual hbridge (2 direction pins and a PWM). Tim's Pro Micro also talks to an OLED screen and his wireless board is put out behind the battery to try to isolate it a little from the metal. The OLED screen needed 3v3 signals, so Tim became a 3v3 logic robot.

He is missing a hub mount, so the wheel on the left is just sitting on the mini gearmotor's shaft. At the other end of the channel is a Tamiya Omni Wheel which tilts the body slightly forward. I've put the battery at the back to try to make sure he doesn't flip over during hard break.

A custom PCB would remove most of the wires on Tim and be more robust. But most of Tim is currently put together from random bits that were available. The channel and beams should have a warning letting you know what can happen once you bolt a few of them together ;)

Tuesday, July 1, 2014

The long road to an Autonomous Vehicle.

Some people play tennis, some try to build autonomous wheeled robots. I do the later and the result has come to be known as “Terry”. In the long road toward that goal what started as a direct drive, “give the motor a PWM of X% of the total power”, lead to having encoder feedback so that the wheel could be turned an exact amount over a given time.  This has meant that the control interface no longer uses a direct power and direction slider but has buttons to perform specific motion tasks. The button with the road icon and a 1 will move the wheels forward a single rotation to advance Terry 6Ï€ inches forwards (6 inch wheels).

The red tinted buttons use the wheel encoders to provide precision movement. Enc left and Enc right turn Terry in place (one wheel forward, one wheel backwards). The Circus and Oval perform patrol like maneuvers. Circus moves forward, turns in place, returns back, and turns around again. It's like a strafing patrol. On the other hand, the Oval turns move like a regular car, with both wheels going forward but one wheel moving slower than the other to create the turning effect. The pan and tilt control the on board camera for a good Terry eye view of the world.

I added two way web socket communications to Terry this week. So now the main battery voltage is updated and the current, err, current for each motor is shown at the bottom of the page. The two white boxes there give version information so you can see if the data is stale or not. I suspect a little async javascript on some model value will be coming so that items on the page will darken as they become stale due to lag on Terry or communications loss.

I've been researching proximity detection and will be integrating a Kinect for longer distance measures soon. Having good reliable distance measurements from Terry to objects is the first step in getting him to create maps of the environment for autonomous navigation.

Wednesday, June 25, 2014

Atmel Atmega1284

I started tinkering with the Atmega1284. Among other things it gives you an expansive 16kb of SRAM, 2 uarts and of course looking at the chip a bunch more IO. A huge plus is that you can get a nice small SMD version and this 40 pin DIP monster with the same 1284. Yay for breadboard prototypers who don't oven bake each board configuration! The angle of photo seems to include the interesting bits. Just ignore the two opamps on the far right :)


I had trouble getting this to work with a ceramic resonator. The two xtal lines are right next to each other with ground just above but the 3 pins on the resonator were always a bit hard to get into the right configuration for these lines. Switching over to a real crystal and 22pF caps I got things to work. The symptoms I was having with the resonator included non reproducibility, sometimes things seemed to upload sometimes not.  Also, make sure the DTR line is going though a cap to the reset pullup resistor. See the wiring just to the right of the 1284.

I haven't adapted the Arudino makefile to work on this yet, so unfortunately I still have to upload programs using the official IDE. I have the makefile compiling for the 1284 but das blinken doesn't work when I "make upload".

Thursday, June 19, 2014

3d printed part repo for attaching to Actobotics structure

I started a github repo for 3d print source files to attach things to Actobotics structure. First up is a mount to secure a rotary encoder directly to some channel as shown below.


As noted in the readme, I found a few issues post print, so had to switch to subtractive modelling (dremel time) to complete it. I leant a few things from this though so the issues are less likely to bite me in the future. There are some comments in the scad file so hopefully it will be useful to others and maybe one day I'll get a PR to update it. The README file in the repo will probably be the one true source for update info and standing issues with each scad file.

You can just see the bolt in below the rotary encoder. That has a little inset in the 3d part to stop it from free spinning when you screw in the 6-32 nut from below the channel.

Sunday, June 15, 2014

Arduino and wireless networking

The nRF24L01 module allows cheap networking for Arduinos. Other than VCC which it wants at 3v3 it uses SPI (4 wires) and a few auxiliary wires, one for an interrupt line. One of the libraries to drive these chips is the RF24. Packets can be up to 32 bytes in length, and by default attempts to send less than 32 bytes result in sending a whole 32 byte block. Attempts to send more than 32 bytes seem to result in only the first 32 being sent. There is a CRC which is a 2 byte and is on by default.

For doing some IoT things using a HMAC might be much better choice than just CRC. The major advantage being that one can tell that something coming over a wireless link was from the expected source. Of course, if somebody has physical access to the arduinos then they can clone the HMAC key, but one has to define what they want from the system. The HMAC provides a fairly good assurance that sensor data is coming from the arduino you think it is coming from rather than somebody trying to send fake data. Another rather cute benefit is that the system doesn't accept junk data attacks. If your HMAC doesn't match the packet doesn't get evaluated by the higher level software.

Using a Sha256 HMAC the return ping times go from 50 to 150ms. The doubling can be explained by the network traffic, as the HMAC is 32 bytes and will double the amount of traffic. I think perhaps the extra 50ms goes into hashing but I'll have to measure that more specifically to find out.

Once I clean up the code a little I'll probably push it to github. A new RF24HMAC class delegates to the existing RF24 class and sends a HMAC packet right after the user data packet for you. The interface is similar, I'm adding some writenum() calls which I might make more like nodejs buffer. Once you have added the user data packet call done() to send everything including the HMAC packet.

RF24 radio(9,10);
RF24HMAC radiomac( radio, "wonderful key" );
radiomac.beginWrite();
radiomac.writeu32( time );
bool ok = radiomac.done();


The receiving end boils down to getting a "packet" which is really just the 32 bytes of user data that was sent. The one readAuthenticatedPacket() call actually gets 2 packets off the wire, the user data packet and the HMAC packet. If the hmac calculated locally for the user data packet does not match the HMAC that was received from the other end then you get a null pointer back from readAuthenticatedPacket(). The data is either authenticated or you don't get to see any of it.

RF24HMAC radiomac( radio, "wonderful key" );
uint8_t* packetData = 0;
if( packetData = radiomac.readAuthenticatedPacket() )
{
     int di = 0;
     uint32_t v = readu32( packetData, di );
     got_time = v;
     printf("Got payload %lu...\n\r",got_time);
}


Oh yeah, I also found this rather cute code to output hex while sniffing around.

Monday, June 2, 2014

TerryTorial: Wheel Feedback, Object detection, Larger size, and now a screen!

Terry the robot now has grown up a little, both semantically and physically. The 12 inch channel that was the link to the rear swivel wheel now has many friends which are up to 15 inches higher than the base beam. This places Terry's main pan/tilt camera at his top a little bit above table height, so he is far more of a presence than he used to be. On the upside, the wheel controller, shaft encoders, and batteries can all live on the lower level and there is more room for the actual core of Terry up top. I'm likely to add one or two more shelves to the middle of Terry for more controllers etc.


There is now also a 16x32 RGB matrix screen up front so Terry can tell you what he is thinking, what speed his wheels are rotating at, and if there is an obstacle that has been detected.

I'm rendering the framebuffer for the screen on the BeagleBone Black using Cairo and Pango (thus & freetype). The image data is extracted from Cairo as 32bit RGBA and packed down into planar data that the arduino expects. That packed framebuffer is then sent over UART to an Arduino 328 which takes care of refreshing the RGB matrix so that fake colour levels are achieved using a software PWM/BAM implementation. Yay for the BBB having 4.5 UARTs. I need to work out how to bring up the TX only UART as that is perfect for the one way communication used to drive the screen.

I should be able to get better colour precision using a Teensy 3.x as the screen driver. The Cortex-M4 just has more cycles to be able to softpwm the screen faster to be able to get a greater perceived colour range.

I customized the font a little in FontForge. Specifically I modified the kerning for "Te" to not waste as much space. The font is based on Cantarell by Dave Crossland. I've only just started on this open font to LED matrix stuff, but with some form of PWM and FreeType rendering the data I hope to be able to get nice antialiased font renders, even at the low resolution of 16x32.

The wheel encoders make a huge difference to the whole experience. Even without full autonomy the encoders allow you to have a repeatable oval or patrol path which can be followed. Also being able to setup the speed to avoid large acceleration or jerky movement so that the higher Terry is still a very stable Terry when on the move.

Monday, May 26, 2014

To beep or not to beep.

There comes a time in the life of any semi autonomous robot when the study of the classics occurs. For Terry, the BeagleBone Black driven Actobotics construction, short of visiting the Globe the answer would seem to be at current "To Beep".



Terry is now run by a RoboClaw 2x5A controller board with two 1024 pulse per revolution shaft encoders attached to an 8:1 ratio gearing on the wheel. This gives an overall of about 13 bits of precision from the shaft encoders relative to the wheel. In the future I can use this magnificent 4 inch alloy gear wheel as a torque multiplier by moving the motor to its own tooth pinion gear. Terry now also has an IR distance sensor out front, and given the 512mb of RAM on the BeagleBone some fairly interesting mapping can be built up by just driving around and storing readings. Distance + accurate shaft encoders for the win?

To use the RoboClaw from the BeagleBone Black I created a little nodejs class. It's not fully complete yet, but it is already somewhat useful. I found creating the nodejs interesting because when coding for MCUs one gets quite used to the low level style of sending bytes and stalling until the reply comes in. For nodejs such a style doesn't implement well. Normally one would see something like

doMagic( 7, rabbits, function() {
    console.log("got them rabbitses, we did");
});

Where the callback function is performed when the task of pulling the number of rabbits from the hat is complete. For using the RoboClaw you might have something like

claw.getVersion( function( v ) {
  console.log("version:" + v.value );
});
claw.setQPPS( 1024 );

Which all seems fairly innocent. The call to getVersion will send the VERSION command to the roboclaw with a given address (you can have many claws on the same bus). The claw should then give back a version string and a checksum byte. The checksum can be stripped off and verified by the nodejs.

The trouble is that the second call, to set of the quadratic decoder pulses per second, will start to happen before the RoboClaw got a chance to service the getVersion call. Trying to write a second command before you have the full reply from the first is a recipe for disaster. So the nodejs RoboClaw class maintains a queue of requests. The setQPPS() is not run right away, but enqueued for later. Once the bytes come back over the UART in response to the getVersion() then the callback is run and the RoboClaw nodejs class then picks the next command (setQPPS) and sends that to the RoboClaw. This way you get the strict ordered serial IO that the RoboClaw hardware is needing, but the nodejs can execute many commands without any stalling. Each command is handled in the order it is written in the nodejs source code, it just doesn't execute right away.

Next up is trying to tune the PID controller variables for the robot. One might expect a command such as "move 1 wheel circumference forward" to work fairly exactly if the quality of the shaft encoders is good. Though for the wheel size and QPPS I get a little bit of overshoot currently. I suspect the current D is not sufficient to pull it up in time. It may be a non issue when the wheels are down and friction helps the slow down process.

Terry now also has an IMU onboard. I created a small class to read from the TWI HMC5883 magneto. You'll have to figure your declination and I have a small hack on about line 100 to alter the heading so that a 0 degree reading means that physically Terry is facing North. I should bring that parameter out of the class so that its easier to adjust for other robots that might mount their IMU in a different physically orientation than what I happened to use.

Saturday, May 10, 2014

Going from Arduino TWI to BeagleBone Black Bonescript

It's handy for a robot to know what orientation it is holding, and for that a Magnetometer can give you where north is at relative to the chip orientation. For this I used the HMC5883. I started out with a cut down minimal version of code running on Arduino and then started porting that over to Bonescript for the BeagleBone Black to execute.

The HMC5883 only needs power (3v3), ground, and the two wires of a TWI.

One of the larger API changes of the port is the Arduino line:

Wire.requestFrom( HMC5883_ADDRESS_MAG, 6 );

Which takes the TWI address of the chip you want 6 bytes from. It has to be prefixed by another TWI transmission to set the read byte address to start from where you would like, in this case HMC5883_REGISTER_MAG_OUT_X_H_M = 0x03.

On the Bonescript side this becomes the following. The TWI address isn't used because the wire object already knows that, but you pass the register address to start reading from directly to the request to read bytes:


self.wire.readBytes( HMC5883_REGISTER_MAG_OUT_X_H_M,  
BUFFER_SIZE, 
function(err, res) { ... });

I pushed the code up to github bbbMagnetometerHMC5883. Note that getting "north" requires both a proper declination angle for your location and also some modification to allow for how your magneto chip is physically oriented on your robot. Actual use of the chip is really easy with the async style of nodejs.

Tuesday, April 29, 2014

Actobotics and ServoCity, addictive Robot fun!

I have had the great fortune to play around with some of the Actobotics components. The main downside I've seen so far is that its highly addictive!


What started out as a 3 wheel robot base then obtained a Web interface and an on board access point to control the speed and steering of the beast. The access point runs openWRT (now) and so with the Beagle there are two Linux machines on board!

The pan and tilt is also controlled via the Web interface and SVGA video streams over the access point from the top mounted web camera. There is a Lithium battery mounted in channel below the BeagleBone Black which powers it, the camera, and the TPLink access point. That block of 8 AA rechargables is currently tasked with turning the motors and servo.

It's early days yet for the robot. With a gyro I can make the pan gearmotor software limited so it doesn't wrap the cables around by turning multiple revolutions. Some feedback from the wheel gearmotors, location awareness and ultrasound etc will start to make commands like "go to the kitchen" more of a possibility.

Sunday, March 2, 2014

The vs1063 untethered

Late last year I started playing with the vs1063 DSP chip, making a little Audio Player. Quite a lot of tinkering is available for in/output on such a project, so it can be an interesting way to play around with various electronics stuff without it feeling like 10 das blinken lights tutorials in a row.

Over this time I tried using a 5 way joystick for input as well as some more unconventional mechanisms. The Blackberry Trackballer from SparkFun is my favourite primary input device for the player. Most of the right hand side board is to make using that trackballer simpler, boiling it down to a nice I2C device without the need for tight timing or 4 interrupt lines on the Arduino.




The battery in the middle area of the screen is the single 16850 protected battery cell that runs the whole show. The battery leads via a switch to a 3.7v -> 5v step up to provide a clean power input. Mid way down the right middle board is a low dropout 3v3 regulator from which the bulk of the board is running. The SFE OLED character display wants its 5v, and the vs1063 breakout board, for whatever reason, wants to regulate its own 3v3 from a 5v input. Those are the two 5v holdouts on the board.

Last blog post I was still using a Uno as the primary microcontroller. I also got rid of that Uno and moved to an on board 328 running at 8Mhz and 3v3. Another interesting leaning experience was finding when something 'felt' like it needed a cap. At times the humble cap combo is a great way to get things going again after changing a little part of the board. After the last clean up it should all fit onto 3 boards again, so might then also fit back into the red box. Feels a bit sad not to have broken the red box threshold anymore :|

Loosely the board comes out somewhat like the below... there are things missing from the fritzing layout, but its a good general impression. The trackballer only needs power, gnd, I2C, and MR pin connections. With reasonable use of SMD components the whole thing should shrivel down to the size of the OLED PCB.

Without tuning the code to sleep the 328 or turn off the attin84 + OLED screen (oled is only set all black), it uses about 65mA while playing. I'm running the attiny84 and OLED from an output pin on the 4050 level shifter, so I can drop power to them completely if I want. I can expect above 40hrs continuous play without doing any power optimization. So its all going to improve from there ;)

Tuesday, February 4, 2014

attiny screen driver for parallel controlled character displays

This is the "details" post for controlling a 7 bit parallel OLED character display using an attiny as a display driver. See the previous post for an overview of the setup and video.

And now... the code. Apologies for some of the names of directories. Instead of branching and whatnot in git I just copied the code to library(n+1) and added the next feature in line as I went. I might at some stage do a write up detailing the stepping stones to get from the bare attiny84 to the code below.

To make it simpler, I'll call the "main" arduino the Uno. That is the one what wants to run the show. The screen is attached to the attiny84 which I'll just call the tiny84.

This is the code that one uses on the Uno: oled_clientspireal.ino It is designed to look very close to the example it was based on (linked in the second line of its header comment).

The attiny84 runs this attiny_oled_server_spi2.ino The heavy lifting is done in SimpleMessagePassing5 which I link next. The main part of loop() is to process each full message that we get from SimpleMessagePassing5. There is a timeout() logic in there after that which allows the tiny84 to somewhat turn off the screen after a period of no activity on the SPI bus. noDisplay() doesn't turn off the OLED internal power, so you are only down to about 6mA there.
The SimpleMessagePassing5 does two main things (coupling SPI into message byte boundaries in a single file, the academics would be unhappy) SimpleMessagePassing5.cpp Earlier version of SMP did
USICR |= (1 << USIOIE); 
in init(). That line turns on the ISR for USI SPI overflow. But that is itself now done in a pin change ISR so that the USI is only active when the attiny84 has been chip selected.

serviceInput() is a fairly basic state machine, mopping up bytes until we have a whole "message" from the SPI master. If there is a complete message available then takeMessage() will return true. It is then up to the caller to do smp.buffer().get() to actually grab and disregard the bytes that comprise a message.

I'm guessing that at some stage I'd add a "Message" class that takeMessage could then return. The trick will be to do that without needing to copy from smp.buffer() so that sram is kept fairly low.
The Uno uses the shim class Shim_CharacterOLEDSPI2.cpp which just passes the commands along to the attiny84 for execution.
This all seems to smell a bit like it wants to use something like Google protocol buffers for marshaling but the hand crafted code works :) In some ways using GPB for such a simple interface as CharacterOLEDSPI2 might be well be considered over engineering, but I'm still tempted.

Wednesday, January 29, 2014

Ripple counting trackballer hall effects

Sparkfun sells a breakout with a blackberry trackballer on it and 4 little hall effect sensors. One complete ball rotation generates 11 hall events. So the up, down, left, and right pins will pulse 11 times a rotation. The original thought was to hook those up to DPins on the arduino and use an interrupt that covered that block of pins to count these hall events as they came in. Then in loop() one could know how many events had happened since the last check and work from there. Feeling like doing it more in hardware though, I turned to the 74HC393 which has two 4bit ripple counters in it. Since there were four lines I wanted to count I needed two 393 chips. The output (the count) from the 393 is offered in four lines per count (its a 4 bit counter). So I then took those outputs and fed them into the MCP23x17 pin muxer which has 16 in/out pins on it. I used the I2C version of the MCP chip in this case. So it then boils down to reading the chip when you like and pulsing the MR pin on the 393s to reset the counters.


In the example sketch I pushed to github, I have a small list of choices that you can scroll through and if you stop scrolling for "a while" then it selects the current entry. Which just happens to be the exact use case I am planning to put this hardware to next. Apart from feel good factor, this design should have less chance of missing events if you already have interrupt handlers which themselves might take a while to execute.