:idea: Jaluro!

[air.command]

Hi Tobi,

Sounds great what you are doing. One possible option that would be nice is to set up a very simple low level API to the rover and open up access to other Frednet developers to try various control and navigation strategies with the robot remotely.

If your robot was set up in a sandbox(not necessary to have actual sand in it at this stage) and hooked up wirelessly to a server connected to the internet, developers could develop their code using the simple API and download their software to it automatically and remotely and see how it behaves. You would allocate time slots to the developers for them to test run their algorithms and functionality.

The side benefit of this would be helping to establish the skeleton remoting framework for controlling a distant robot, and doing software updates on the fly. We could easily add the expected lag when dealing with a robot on the moon.

I think there are a lot of FREDNET members that are keen to start cutting code even though we are still a long way off actual flight software, but many valuable lessons can be learned and protocols worked out in the prototype stage.


- George

[tobiokanobi]

That's a great idea Georg, but it will take me until Feb /March to set something like that up. Can't promise it, but even if Jaluro will not make it to Luna (but some other Frednet Rover does) - this might be a way to contribute a bit to out success.

One of the next tasks I wanted to open is the earth control interface for the rover control. I will try to implement your suggestion. I might need help here!

-tobi

[alexc]

There might be something useful at http://surveyor.com/
The surveyor was used in the demo on CNBC back in 2007. I think much of it is open source.

[tristancho]

Good Job, Tobi.

I have added some links for the wiki Solar Power as Tobi has proposed.
http://wiki.xprize.frednet.org/index.php/Power_Solar

Remember that aluminum is a good reflector:
Density: d=2675 kg/m3
Thermal conductivity: 220 W/m·K @ 20ºC
Emissivity e=0.09
... and carbon fiber is just the opposite.
Density: d=1560 kg/m3
Thermal conductivity: 1 W/m·K @ 20ºC
Emissivity e=0.98
http://www.s271082832.e-shop.info/

http://www.infrared-thermography.com/material-1.htm
http://www.engineeringtoolbox.com/emiss ... d_447.html

Thermal team may take in account in the calculations.
Best regards,
Joshua

[AndersFeder]

There might be something useful at http://surveyor.com/
The surveyor was used in the demo on CNBC back in 2007. I think much of it is open source.

That page got me thinking of stereoscopic video from the Moon ... that would be so cool. Too bad we're on a power budget.

[AndersFeder]

Two articles on Jaluro has appeared on the Control Engineering website here and here.

[alexc]

Is there something horribly wrong with the Control Engineering website? I am prompted for password then, after canceling a few times I get a page which looks like the web pages we used to have in 1992...

[AndersFeder]

Is there something horribly wrong with the Control Engineering website? I am prompted for password then, after canceling a few times I get a page which looks like the web pages we used to have in 1992...

It looks fine to me, neatly layed out with CSS and everything. I were asked for a password once though, but not after reloading. Possibly a transient error.

[ryanw]

I'd like to see the '4 pages' of math he was talking about

[alexc]

Jaluro is now also on space.com:
http://www.space.com/businesstechnology ... rover.html

[AndersFeder]

Way to go! Very high-profile site in the online space community.

[air.command]

Jaluro is now also on space.com:
http://www.space.com/businesstechnology ... rover.html

Good to see! I went through the comments today, and while a lot of people mention the low clearance issue, one of the people mentioned the Segway. Which I thought was quite an interesting suggestion.

Perhaps that is something we can look into. Instead of the weight hanging down, it is actively balanced on top. You get the ground clearance you need and provide a higher platform for the camera. The rover could have two caster wheels either side of the platform raised considerably off the ground so that the rover can enter passive stability mode when stationary, by falling over to either side. It could possibly travel in passive mode as well.

In lunar gravity it may require less power to keep it actively balanced.

[Joerg]

Here the new wheel design for JALURO.
A upsized clone of the WRV1 wheels:


http://wiki.xprize.frednet.org/index.php/Image:JALURO-WHL-U.png

Jörg

[tobiokanobi]

First: Thanks Joerg for your wheel drawings!! I appreciate it a lot.

Second: Jaluro is now wireless "off the leach", you can see the video on our Team FREDNET blog or on our Youtube Channel.

There is still a lot to do and even more redesigning neccessary, but the basic concept seems to work.

Third: We established a remote control for Jaluro from Copenhagen to Vienna, with the code published in Jaluros Software Sandbox. Unfortunatlly video recording did not work well, so we can't show the video stream session we had that night.

Thanks to Alex, Anders, Jeyram and Yeonhoo for their great help!!!

-back to work

tobi

[tristancho]

Tobi, congratulations!
Joshua

[ross_brigoli]

Here the new wheel design for JALURO.
A upsized clone of the WRV1 wheels:


http://wiki.xprize.frednet.org/index.php/Image:JALURO-WHL-U.png

Jörg

I suggest making the Jaluro wheels hemisphere(half of a sphere). This will prevent the position of being standing vertically when tipped-off as it stumbles to rocks and or inclined planes.

The rover could stuck in the dusty lunar soil when in vertical standing position.

[tobiokanobi]

Good suggestion Ross - i'll keep it in mind!

-tobi

[AndersFeder]

Clever. While we're at it, I wonder if it would be feasible to make the center of mass of the rover adjustable so you can 'wiggle' it a bit in such situations, and maybe have it fall back in the right orientation.

Picorover clearly has an advantage in these situations though!

[ross_brigoli]

About the rocking problem, I think a PID controller can do the job by automatically adjusting the motor speed with respect to the angle of the axle/body relative to the horizon.

In other words, by making the drive motors react to this very low frequency vibration/oscillation.

http://www.ecircuitcenter.com/Circuits/pid1/pid1.htm

http://en.wikipedia.org/wiki/PID_controller

-Ross

[alexc]

Did you notice in the latest video that Jaluro was rocking much less than during the first tests? Or is it just my imagination?

[tristancho]

About the rocking problem, I think a PID controller can do the job by automatically adjusting the motor speed with respect to the angle of the axle/body relative to the horizon.

You got it!

Picorover uses two PID in order to control Pitch and Roll angle. If Jaluro is able to control pitch angle, it will control proportionally the thrust like Picorover does.

Jaluro is longitudinal static stable and lateral static neutral. As per WRV1, since center of gravity is fixed, no oscillations are produced in the stability model hence it is longitudinal and lateral static neutral. Picorover is lateral and longitudinal static stable.

I found that Team FREDNET we have a good spread of Rover types. I Like.
Joshua

[tristancho]

Hi,
In this video you can see the longitudinal unstable version of Jaluro. I prefer Jaluro because it do not need to spend energy in order to stabilice the longitudinal mode.

Regards,
Joshua

[tobiokanobi]

Thanks for your inputs. Acutally I thought on installing an accelerometer and just measure the vector movements into each direction and (maybe) to adjust the speed accordingly.

Now back to organiztaion : There is a new status report, not big news, but for the record.

Learned/Experience until now:

*) Still the wheels needs to be redesigned. I think I will take Jörgs proposal and adjust it to fit to Jaluro. I might change a bit here and there, but the flex concept will stay

*) Axle needs a lot of improvement

*) Without encoder = hard to drive straight

*) Body shorter and slimmer

*) Commands given by the Server to the Client(Jaluro) don't work all the time, don't know why.

*) Radio distance is way too short

*) A "controlled" de/accelleration lowers the swinging of the body

*) Fine adjustment of the movements will be neccessary

Okay, a lot to do. Most of them I can solve, with a bit of time. We started now in FREDWEBwith the Rover Director. This should hopefully solve some problems with the command transmission.

The rest is hard hard work. Let's start. (Help welcome )

-Tobi

[tristancho]

Good job Tobi!
Joshua

[tobiokanobi]

Unfortunatly I have to suspend "Jaluro!" within FREDNET and will make it to a private project. It was a very very sad and not easy decision.

I appreciate your help and comments a lot, and I want to say "Thank you very very much" to everyone who helped me and Jaluro.

-tobi

[tristancho]

Hey! Tobi,

You have done an impressive work with Jaluro.
Good luck for the private project. Jaluro's development will continue.

Regards,
Joshua

[alexc]

There is now also a prototype of Jaluro for manned lunar missions, see http://www.googlelunarxprize.org/lunar/ ... unar-rover
It doesn't even require electrical power for driving and steering!

[scasey]

What I like about this is that they have published the design here:

http://www.instructables.com/id/The-SteamRoller/

It's another example of an open source hardware development.

SC

[tristancho]

NIce version of Jaluro!
Joshua

[tobiokanobi]

-tobi

P.S Next version release of Jaluro comes hopefully soon

[tristancho]

Can't wait!
Joshua

[tobiokanobi]

Pre release of the new carbon wheel rim - complete rim should be finished tomorrow:

(Used as usally my cheap mobile phone camera - bad quality pictures )





Hope you like it,

Tobi

P.S. in the back you see the old rim made out of Al - a lot heavier and approx. 1cm more diameter (old:20cm, new:19cm)

[tristancho]

Hey Tobi,
Awesome!
This is a very nice job.
Congratulations.
Joshua

[alexc]

That looks promising, Tobi!

[jdp]

nice fab, as usual!

[tobiokanobi]

Thx guys!

Here are now the new pictures, wheels 97% completed (cosmetic surgery still required).

Best of all, through usage of the carbonfiber I saved more then 2/3 of the original wheel weight (design#1 Al: 300g, design#2 Carbon/Al: <100g)

In picture#1 you see also the old wheel, the old body - and the NEW carbon body (still under construction )





Hope you like it.

-Tobi

[tristancho]

Are so nice the wheels in front of the nanoStation2 !
I like very much.
Congratulations again, Tobi.
Joshua

[MarcBrakels]

Nice work

[tobiokanobi]

I did some work for another projects and had to design a simple power board, and since it is very usefull, I will use it for Jaluro! too:



Input voltage range from 9-18VDC
Output:
1x12V, 1250mA
2x5V, each 1000mA
1x3.3V, 400mA
and direct battery power for camera and motor controller (both control itself the input/output voltage).

It's not finished yet, but functionally it is working.

-tobi

[MarcBrakels]

Nice work.

Can you post some diagram of battery power part, I am very interested in that part.

[tristancho]

Hey Tobi,
Very good job and also Open Project.
Regards,
Joshua

[tristancho]

Nice work.

Can you post some diagram of battery power part, I am very interested in that part.

Also size board and weight if possible...
Joshua

[tobiokanobi]

Here a schematic, but there is currently no solar power or charger included (I use it for a different project too).The second picture shows my power source, as I said I prefer nanophosphate batteries and I've choosen "A123".





-tobi

[tobiokanobi]

Late Night Soldering

(Green light indicates power on)



All output values within 1% tolerance.

-tobi

[tobiokanobi]

I had some minutes time, so I did some surgery on a CAT5 cable and powered the camera and the WIFI (Pico2HP from Ubiquiti) station. All power LEDs are on, so I suspect that it works. Tonight I'll do some measuring and configure the Wifi station.



-tobias

[alexc]

Soon time for a new session on ustream
This time I'll have my recording set up and ready!

[tristancho]

Hey Tobi,
Great work!
Joshua

[tobiokanobi]

I took my Volt- and Ampmeter and measured the values:

All switched on (Camera, Picostation, Powerboard) consume at 14.4V about 0.6A.

-tobi

[MarcBrakels]

Don't forget the watch the battery voltage(no lower then 3.0 V per cell|12 V total).

I have broken some cell's this way

[tobiokanobi]

Don't forget the watch the battery voltage(no lower then 3.0 V per cell|12 V total).

I have broken some cell's this way

Thanks for info Marc, I have an eye on this.