:idea: Jaluro!

[tobiokanobi]

Hi to all!

I started a few days ago with Jaluro - Just another lunar rover - and I would warmly invite anyone who wants to - to participate. There still much
to organise, and it will take me one or two more weeks to do it, but anyone who wants to join allready now (or later) is invited to do so! If you're interested, follow this link:

http://wiki.xprize.frednet.org/index.php/Portal:Jaluro

Waiting for your input ,

--tobi

P.S. This is until ?? not an officiall project here at frednet, I will inform Joerg and Fred tomorrow

P.P.S. I'm very busy, so I will not spend more than 1 hour daily onto this, but I do as much as I can - and as good as possible

[tobiokanobi]

Can someone provide a link to the current GLPX Rules?

Thx,

--tobi

[alexc]

I don't think we have v2.0 of the rules uploaded to our servers. We got them by email for the review process. I can send you a copy if you want; however, v3.0 and the MTA (master team agreement) should be out very soon - we expected them already in August.

[tobiokanobi]

Thx alex!

* By the way, do you have any proposals for an Antenna?

* What would you think, is an 500mW Amplifier reasanable to extend the range of the lunar rover (heading for 20km)

--tobi

[tobiokanobi]

Got Joerg's okay, so Jaluro is a bit more on the way to success!



Moved Jaluro today to a portal...and started to organise the whole thing

-tobi

[alexc]

Thx alex!

* By the way, do you have any proposals for an Antenna?

* What would you think, is an 500mW Amplifier reasanable to extend the range of the lunar rover (heading for 20km)

--tobi

Assuming that we'll try off-the-shelf wireless network technology, I think to begin with we should try an omni directional whip on the rover and a directional antenna on the "lander". It seems we can get relatively cheap directional antennas with 10-20dB gain and that is quite significant.

That should work up to a few kilometers. For longer range we need to have a look at this new WiMAX-thingy, see http://en.wikipedia.org/wiki/Wimax. It looks interesting, should be able to provide 10Mbps @ 10km, although the availability of equipment is not quite clear to me. In any case, there needs to be free line of sight; minor obstacles should be ok though.

As it has been pointed out by Daniel earlier, we can expect the Fresnel zone, see http://en.wikipedia.org/wiki/Fresnel_zone, to be a real problem for longer ranges, because our antennas will be very close to the surface. We have to try under realistic conditions, so a prototype rover is very welcome

[AndersFeder]

That should work up to a few kilometers. For longer range we need to have a look at this new WiMAX-thingy, see http://en.wikipedia.org/wiki/Wimax. It looks interesting, should be able to provide 10Mbps @ 10km, although the availability of equipment is not quite clear to me. In any case, there needs to be free line of sight; minor obstacles should be ok though.

For what it's worth, I used WiMAX for my broadband connection about a year ago. Of course there is no trees or rooftops on the Moon, but it was very difficult to get a connection in my case. I had to place the transceiver way out on the lawn and even in some areas there the connection would be broken. When I did find a spot with a clear connection, it worked flawlessly though.

I still have the transceiver unit - if it's of any use, I would happily donate it to the lab. I don't have any kind of specifications for it however (its a standard 'black box' acquired from the ISP).

[tobiokanobi]

I still have the transceiver unit - if it's of any use, I would happily donate it to the lab. I don't have any kind of specifications for it however (its a standard 'black box' acquired from the ISP).

I will come back on the donation soon! Thanks!

[tobiokanobi]

Can someone please review the follwing topics (errors, missing/wrong content...)

http://wiki.xprize.frednet.org/index.ph ... o_Analysis

http://wiki.xprize.frednet.org/index.ph ... Objectives

http://wiki.xprize.frednet.org/index.ph ... ganisation

If you find something that is missing, please add it or post the missing information here.

Thanks,--

tobi

[nenea dani]

That should work up to a few kilometers. For longer range we need to have a look at this new WiMAX-thingy, see http://en.wikipedia.org/wiki/Wimax. It looks interesting, should be able to provide 10Mbps @ 10km, although the availability of equipment is not quite clear to me. In any case, there needs to be free line of sight; minor obstacles should be ok though.

We have to try under realistic conditions, so a prototype rover is very welcome [/quote]

It is time to build a robot working very simple to test the wireless connection and the radio link proposed by Alex and why not to see how it moves in the field. I thought at a very cheap option, the SSC-32 Servo Controller from http://www.lynxmotion.com/Product.aspx? ... egoryID=52 with SSC-32 Sequencer (seq-01 ). The control is on serial port, but I seen in the forum that it can be used on USB with a DB-9 converter (the lynxmotion forum).
My idea is to add a DB9-RJ45 converter and connect it in a switch wich will be connected at a Wi-Fi device. In the other 2 ports of the switch we will connect the both Megapixel IP cameras or only one for test.
SSC-32 Servo Controller has a lot of facilities, including combinations of servo. The most important advantage is that we can use four outputs in order to control the 4 brushless syncronised motors used for traction/drive . We can easily use a complete kit with brushed motors/encoders and gearboxes like this one. http://www.lynxmotion.com/Category.aspx?CategoryID=111
However, it takes a lot of servos. The complicated problem is that a group of software specialists to integrate the movment control of the rover in a software that will show the images sent, the movment and the feedback signals. SSC-32 Sequencer is not open source and I don’t know what can be made. If this problem is solved we can use another Rover computer for the rest of the functions or wich will have all the things in one. The communication idea USRP of Alex I consider it the perfect with our project and I think that is must be backed.
A Wi-Max link on 20km is hard to belive that can be made on Earth with 2 antennas up to 1m height and on the Moon it cannot be made because the curvature that is bigger. Wi-Max has 2 licensed bands 3,3 – 3,8 GHz & 2,3 – 2,7 GHz and one unlicensed 5,725 – 5,85 GHz.

[ryanw]

if you guys are trying to save mass, i suggest using this servo controller:
http://www.pololu.com/catalog/product/207
its only a few grams(compared to 73) , we use it on the hover test vehicle to control servo actuated throttle valves.

[tobiokanobi]

Give us a bit more time - please don't hesitate - a very basic rover for testing should be set up until January.

-tobi

P.S. It is not clear if we go for wifi/wimax - Joerg proposes a half duplex solution -> but nothing is fix until it is build!

[alexc]

Good work on the wiki, Tobi.
The wifi and wimax are just on the "to-be-investigated" list because they are off the shelf technologies that can provide high data rates required by HD video. In the real mission this can turn out to be a great advantage because we might save weight and power by doing less video processing on the rover.
But if you can do with something else for the rover prototype - it's entirely up to you

[tobiokanobi]

Hi Anders

I still have the transceiver unit - if it's of any use, I would happily donate it to the lab. I don't have any kind of specifications for it however (its a standard 'black box' acquired from the ISP).

I suspect it's a usb dongle wimax transceiver or? Do you have a brand or type,...

Thx

tobi

[tobiokanobi]

Just out of curiosity...nobody of you has a wimax basestation somewhere in his/her backyard/storage, waiting for getting donated?

-tobi

[AndersFeder]

I suspect it's a usb dongle wimax transceiver or? Do you have a brand or type,...

The box has an RJ-45 ethernet interface. It's rebranded by Clearwire, the WiMAX ISP. I have looked around the net but I can't immediately find any sort of technical specifications. All I've got is this non-technical Danish/English user manual.

[tobiokanobi]

Jaluro is grown a bit, check the specification out:

http://wiki.xprize.frednet.org/index.ph ... ifications

I know, everything is still a bit wide spreaded but I have to start somewhere.

By the way, phuzzy3D will make some drawings, coming soon (I've to send him bit more info)

@anders: thx for info ...and can you send me your eMail address?

Has anyone of you something like this:

http://www.tri-m.com/products/engineering/vsx104+.html

in his shelf or backyard and is willing to donate (for testing purpose)?

See ya on moon,...

tobi

[tobiokanobi]

If some is good in model making, this can be a task for you:

http://wiki.xprize.frednet.org/index.php/Jaluro_Tasks


see ya on moon

-tobi

[tobiokanobi]

There are some new part drafts of Jaluro out,

http://wiki.xprize.frednet.org/index.ph ... o_Drawings

Have a look!


see ya on moon

[alexc]

Cool! Let me know when you think there is a complete set and I'll post them on our X Prize blog.

[air.command]

Hi to all!

I started a few days ago with Jaluro - Just another lunar rover - and I would warmly invite anyone who wants to - to participate. There still much
to organise, and it will take me one or two more weeks to do it, but anyone who wants to join allready now (or later) is invited to do so! If you're interested, follow this link:

http://wiki.xprize.frednet.org/index.php/Portal:Jaluro

Waiting for your input ,

--tobi

Hi Tobi,

Seems like you have put quite a bit of effort into this. Excuse me if you have already addressed this, but have you considered the following in your design:

- Ground clearance. What will Jaluro do when you sit it on lunar soil that is 1inch of fine powder? The wheels are fairly narrow and will tend to sink more into the soil. Especially if there is any slippage. The body of the rover will likely beach itself in this case.

- What is the greatest obstacle that a wheel can climb over, given the limited torque arm?

- The wiki mentions solar panels but I don't see any in the design. Is the body covered with these?

[tobiokanobi]

Hi aircommand!

Thanks for your questions, I might not be able to give a 100% answer, but I try my luck:

- Ground clearance. What will Jaluro do when you sit it on lunar soil that is 1inch of fine powder? The wheels are fairly narrow and will tend to sink more into the soil. Especially if there is any slippage. The body of the rover will likely beach itself in this case.

I know that Jaluro has a very limited ground clearance - we could increase it if we go for bigger wheels - but this takes more space, and increases weight. We have to look how it work best. The rover will be (most probably) very light, and if the "touch" area of the wheel is sufficient big (again depending on the wheel dimensions), it hopefully will not sink to far into the regolith. Testing this will a good advice and will show it the rover will beach itself. Did you have a look on the lunokhod? This rover looks a lot heavier then my design.

- What is the greatest obstacle that a wheel can climb over, given the limited torque arm?

A good question. It will depend on the weight (on luna) of the rover body. and the proportion of the wheel radius to the "offset" of the body center axis to the wheel center axis. We might be able to increase it further if we can lower the center of gravity of the body. I don't have right now now idea how big the final weight will be nor of the offset. Everything is open.

Certainly, it will be a bad rock climber, but it should be able to climb hills. (I've added in the organisation section an external link to a similair 2 wheel robot - so have a look)

- The wiki mentions solar panels but I don't see any in the design. Is the body covered with these?

My fault- I didn't include them in the drawing. But yes, the upper half of the rover body will be coverd with solar panels. Except Cameras

I know that the answer are a bit impresice and I know that the concept has several disadvatages (and also advantages) - but I wanted to start something and not to discuss everything endless.

The Rover itself is the most simplest thing on the whole GLXP Project. And even if I experience at some point that my design isn't the "go" for Frednet, I hope that I can active participate and develop some little things.

I invite everyone to join.

see you on luna,

tobi

[tobiokanobi]

Hi!

Jaluro is looking for the following components (or equivalent):

PC104+ form factor

SBC: VSX104+
www.tri-m.com/products/engineering/vsx104+.html

WiFi module: AIRLAN2000 or WLAN17202ER
www.rtd.com/pc104/um/network/wlan17202.htm
www.ampltd.com/prod/airlan2k.html

Power Supply: HE104 (power can be less, down to 25W)
www.tri-m.com/products/engineering/he104.html

Motor Controller: ESC629
www.rtd.com/pc104/um/controller/esc629.htm

If someone has something like that and does not need it right now and is willing to sponsor it for testing purpose please contact me.

Components can get damaged during testing!

See you on moon,

-tobi

[tobiokanobi]

Jaluro needs someone who takes care of the solar power. What we need is:

someone who is interested into solar power
some calculations if we can gain sufficient power with the existing area of solar cells
what voltage and current we can expect
decides which solar panel we shall take

We will have an approximate area covered with solarcells of: 250 x 150mm. We need approximatly : 20-25W. The output voltage should be between 6 and 40 VDC (best arround 12-15V).

Perhaps a good link http://www.spectrolab.com

Interested? Contact me!

-tobi

[tobiokanobi]

Since no one wanted to do the mock up, I had to start by myself...not finished yet but some parts are allready on the best way ...

I don't spend much time on niceties, so excuse that the pieces are not accurate worked - I've taken the pictures with my mobile phone cam - so very poor quality

Here an example of an intermittend state of one rim, the rest is in the jaluro wiki!

http://wiki.frednet.org/index.php/Jaluro_Drawings



see you on moon,

tobi

[tobiokanobi]

Does anyone has two pieces of an aluminium "duct" with the following dimensions:

Outer diameter: 200mm
Inner diameter: 190mm
Wall thickness: 5mm
Length: between 30 and 40 mm
2 Pieces

Material: AlMgSi0,5, AlMgCuPb or some similar


Thx,


Tobi

[ryanw]

Looking good Tobi,
i'll look around and see if we have anything like that.
-ryan

[WPomerantz]

Just wanted to chime in with a word of encouragement. This is really cool stuff! Keep up the good work, and we look forward to seeing a lot more progress both from the JALURO subgroup and from FREDNET as a whole in the near future!

Best,
Will

[tobiokanobi]

Got these little gearmotors today



-Tobi

P.S. NASA used Maxonmotors and gears during the phoenix mission on mars

[alexc]

Hey, I thought you are only going to need two wheels

[tobiokanobi]

Hey, I thought you are only going to need two wheels

I'm thinking about going for production run

Or see it like this, let's build two jaluro's interlink them through an flexible vertical and horizontal joint - and voilá - WRV1

Ebay

Here are the Specifications and Drawings.

BTW: The Motors are only for the earth-prototype

-tobi

[alexc]

Good idea, build at least two of everything. If the launcher blows up we are ready with #2

[air.command]

Or see it like this, let's build two jaluro's interlink them through an flexible vertical and horizontal joint - and voilá - WRV1

That sounds like a good plan. Redundancy is the order of the day. Will one side be able to pull the other along if it fails completely? Perhaps some clutch mechanism that releases the wheels from the gearboxes so they can turn freely if one side of the rover fails. Disengaging the clutch on one side when driving down slopes can conserve power.

[AndersFeder]

That sounds like a good plan. Redundancy is the order of the day. Will one side be able to pull the other along if it fails completely? Perhaps some clutch mechanism that releases the wheels from the gearboxes so they can turn freely if one side of the rover fails. Disengaging the clutch on one side when driving down slopes can conserve power.

Lunokhod actually had pyros on eight six wheels so it could just blast them off if it should become necessary.

[Humble]

Well, seems that we have a prototype out there trying out the concept ...

http://www.xs4all.nl/~sbolt/e-smiley.html

[tobiokanobi]

Good info and nice to know, "my" concept seems to work. I like the

Hopefully I get Jaluro to drive before Chrismas

-tobi

[alexc]

Sounds great! Whenever it starts moving, don't forget to record on video

[tobiokanobi]

Here some impressions of the current work

Mod a CAT5 cable for powering the elphel cam with an external power source instead of PoE:







Soldering a RS232 to I2C adapter





Testing first time motor controller and driver





Have a nice weekend

Tobi

[tobiokanobi]

I finished today the Jaluro Mock Up / Prototype v0.1

I've taken a video for the first test drive (drive by wire), but unfortunatly, my camcorder supports firewire, but my new laptop does not. I release now a few photos and I will try to take tomorrow a new video, perhaps with the elphel cam

My son helps me out:



And finally finished/assembled





Driven by wire:



BTW: It works - not perfect, since the little control software I've written sucks, everything is still a bit "loose" and finally because it's still controlled and powered via wire. But I can drive each wheel independent, fwd and rev, "speed control" (not really since the encoder is missing, so I just control the pulses to the motor, the encoder comes next time).

If I've sufficient time, the video comes tomorrow - as I promised before Chrismas.

-tobi

[AndersFeder]

Very cool Tobias. Remember to get some good shots for the media too

[tobiokanobi]

Finally, and before Christmas, here a short video on youtube:

http://www.youtube.com/watch?v=lNuIb4_eS-U

-merry christmas and nice holidays

-tobi

P.S: Anders: is the quality of this shot good enough for media?

[AndersFeder]

Yeah, I think it will be okay. But just as a thing to keep in mind, for still images I think they generally like relatively high-res so they can scale to whatever format fits their medium. It's also good to get a few shots from angles that are not 'straight on' - instead of one image of the front, one image of the top and one image of the side, they would rather include a single image showing a little of the front, a little of the side, and a little of the top, say.

Nice video btw I don't know how far ahead the other teams are, but they at least do not seem to be capable of documenting their progress as well as Team FREDNET does.

[tobiokanobi]

Here a short summary of my personal impressions after the first test drive:

* Wheels are to heavy - they need to be a lot lighter. Most of the power of the motors gets lost to turn wheels.

* Casing: I guess I can design the next casing at least 1/3 smaller - most probably I will decrease the heights - so more surface for solar panels.

* Steering: I had no encoders installed - a must to drive a straight path. I see only induction encoders as a possiblity (or generator power?). Optical encoders might get covered by regolith...

* Swinging: During acceleration and decceleration the inertial forces let the body swing fwd/aft. That might be a handicap for video. Dan Smith suggested a dragging glider/roller to avoid swinging. This would be a very easy solution, but I will not integrate it right now - maybe in a later version?

* Control Commands: Currently I send one command after the next one to the rover - (eg. first CW/CCW, then: speed "150", then left wheel). For future control, the commands needs to be bundled and executed "simultanisously" - e.g for a left turn, 5 seconds:
Command pack 1 is:
(RW, CCW, 150; LW, CCW,100; 5s)
Send (Pack 1) to rover
Get feedback of command pack (recieved, checksum,...)
Send execute command
Start watchdog on rover (to interrupt movement in case of fail function)
Get confirmation of accomplished movement


* I need to get rid of the cables

Moreover, it was very easy to get it to drive, hardest thing was to get the wheels. It is very flexible and easy to control.

Has someone a good idea for the wheel design - I've one in mind, but I'm open for any proposal - I'm also thinking about using composite materials? - It will be a hard job to top Jörgs design - but I think I stay more simple.

I'll open soon launchpad soon for the rover control software - either C or C++.

Currently I'm working on the wifi connection, I need to get wifi dongle running and then I should be able to do first cross country tests. Hopefully until end of Jan. Then I will integrate the elphel cam. Shouldn't take me more then one month - maybe earlier.

So I hope to start with fineadjustment and getting closer to a nice looking and full featured rover version soon. If someone whats to help - help is welcome -

Some tasks I will open in our wiki as soon I've a better idea about I what we need:

a power/motor/heater/servo controller with an USB1.1/2.0 interface
a wifimodule with USB2.0, atheros chipset, 250mW output - can be off the shelf too, as long all drivers are opensource and it's possible to get them to run under linux.

Sounds interesting to you? Then contact me.

-happy new year to you all

tobi

[Joerg]

Why not a wheel like this:

http://wiki.xprize.frednet.org/index.ph ... WHL-J1.png

Parts from the WRV1 whell to form a 200mm JALURO wheel.

Saluti Jörg

[tobiokanobi]

Thanks Jörg!

As I wrote - it will be hard to top your wheel design I will consider it seriously.

-tobi

[air.command]

Why not a wheel like this:

http://wiki.xprize.frednet.org/index.ph ... WHL-J1.png

Parts from the WRV1 whell to form a 200mm JALURO wheel.

Saluti Jörg

Hi Jörg, I've been following your wheel development. Have you assembled it for the WRV1 yet with the spokes? I was just wondering what sort of sheer deflection you get with the tread when the robot is sideways on a slope. With the larger wheel diameter and longer spokes it may be an issue for Jaluro. ... just an observation

[Joerg]

@ air.command


Yes the spokes the prototype is ready; see:

http://wiki.xprize.frednet.org/index.php/Portal:WRV1

The WRV1 wheel is very stiff, but with only one wheel i could not do any serious tests. The four (slightly modified) wheels for the tests are in production right now. Since the spokes and the segments are the same i think the JALURO wheel should work, but only a test can show that.

Saluti Jörg

[alexc]

Thanks Jörg!

As I wrote - it will be hard to top your wheel design I will consider it seriously.

-tobi

Maybe Jörg is offering to manufacture two wheels for Jaluro now that he is already making 4 wheels...

That wheel design is really cool and I hope they will work our in practice. Jörg could maybe make a business out of it!

[air.command]

@ air.command


Yes the spokes the prototype is ready; see:

http://wiki.xprize.frednet.org/index.php/Portal:WRV1

The WRV1 wheel is very stiff, but with only one wheel i could not do any serious tests. The four (slightly modified) wheels for the tests are in production right now. Since the spokes and the segments are the same i think the JALURO wheel should work, but only a test can show that.

Saluti Jörg

Ahh excellent, Sorry I hadn't seen the last set of photos with the spokes. As alexc said, the wheel looks really good.

[tobiokanobi]

Just a little info about current work:

*) I've implemented a wireless connection (wifi 802.1g) through a PC104 board
*) I'm working on the rover control software to make Jaluro wireless.

Laptop (earth) is client
Jaluro is working as server

*) I need a bit of time to work me into the communciation with serial/RS232 devices (motor drive unit) in C/C++.
*) I've also implemented the Elphel camera into this system.

Currently almost everything works fine, I can contact the camera, recieve JPEGs and control the camera, but for unknown reason, the gateway (the PC104 board running with debian 2.4 Kernel) blocks rtps data streams through port 554 - I might have set somewhere a wrong broadcast address, flags or something. I haven't installed any filter or firewalls Might be because of the udp protocol - I will try it via TCP and Unicast instead of Multicast and udp.

*) I'm thinking about to use for the next Jaluro Carbon instead of Al.
*) Solar Panels (this concerns also Power System of Lander and Bus), Speedator gave me http://www.azurspace.com/ as a hint.

That's it. Any questions?

-Tobi