Fullscreen
Loading...
Blog: Danny Staple
Description: Thoughts, news, events, creations and insights about Lego, Robotics and technology from the creator of OrionRobots.
Created by orion on Sun 05 of Dec, 2004 12:42 GMT
Last post Wed 03 of Feb, 2010 22:04 GMT
RSS feed (182 Posts | 483875 Visits | Activity=2.00)

The NXT-A-Sketch

Posted by orion orion on Sat 21 of Jul, 2007 16:03 BST
This is somehow really silly, but I have to admire them for the ingenuity. Some person has actually programmed an NXT to behave like an Etch-A-Sketch . Not sure how the shaking to clear works, but it is certainly an interesting use for it.

I might build one myself for the giggle of it. It has set me wandering if I could build some games - use two motors to power a self-centering, force feedback joystick, and then have some silly graphics on the screen. I might start with a worm-eating-pills sort of game.


WARNING: No such module TAGS! NXT Etch-a-sketch

Permalink 0 comments Print Email This Post

Alan Johnston freed.

Posted by orion orion on Fri 06 of Jul, 2007 21:05 BST
I know I am a bit late off the mark here, and it may be off topic, but a note to say that I am very happy to hear that this man has been allowed to continue his life. No doubt the memory of his ordeal will always be with him, but I have great respect for him for going through this, and I hope he will have ample time to spend with his family and getting his life back.

The cowards that kidnapped him, and the cowards that continue to kidnap many other, maybe not so high profile people, are no more than common criminals, and any decent person is probably disgusted with this behaviour. It is one thing to beset, and act against a government or campaign peacefully against an opposing body. It is another to kidnap, maim and threaten to further goals.

Anyway, I salute you Alan Johnston.


Permalink 0 comments Print Email This Post

Prototype Space Hotel Launches!

Posted by orion orion on Fri 29 of Jun, 2007 21:51 BST
Robert Bigelow, a billionaire has launched a prototype space hotel, named Genesis II, into orbit via a Russian craft. It is basically a large inflatable unit, with communications and sensor clusters. Following successful positioning, the hotel is now communicating, sending back images.

The predecessor, Genesis I was launched in July of 2006, and was also a success. As far as I know, it is still in orbit.

The experimental hotel is paving the way for its successor, already named the Nautilus according to engadget, or possibly the Galaxy followed by the SunDancer?. Bigelow Aerospace has even opened up the floor for innovators to get in on making this viable with "America's Space Prize" where $50,000,000 is offered for building a craft capable of carrying a crew of five to 250km and docking with a Bigelow Aerospace module. Unfortunately - this is available only to Americans.

What does the space hotel mean?

This is a pretty interesting development. A space hotel may have implications for the future.

Hopefully this area may allow development of cheaper travelling methods, and push forward the space elevator project, as well as biodome systems. Including botanic gardens on such hotels, as well as considering power generation, food growing, oxygen recycling and so on would be a huge boost to the space program, and even contribute to the viability/survivability of humankind as a whole.

Systems that can effectively process waste (and not merely launch it away) may need to be developed, pushing money into closing loops of consumption, which would help recycling efforts back here on earth.

What Next?

I am interested in when we will see off-planet manufacturing/fabrication plants.

The advantages of an off planet fab plant are many. Contamination Free Zones would be very easy. The scale of the thing would be limited to shipping things up there, but space tourism will make this cheaper.

Since real estate is pretty cheap out there, expansion gives few costs in those terms.

With limited gravity, very large structures or machines could be constructed, as well as fabrication systems that take advantage of weightlessness by projecting things at slower speeds.

Solar power, as well as atomic power systems can be used in ways that are either not so efficient or safe on earth.

Hopefully such systems will not need to be manned, as a downside would be very cheap labourers sent off-world to such labs.

Cost

However, all of this comes at a cost. Rockets burn a huge amount of fuel, and this all ends up adding to the carbon footprints. Systems must be engineered to maintain stable orbits or occupy midpoints with few gravitational influences.


WARNING: No such module TAGS! space contest

Permalink 0 comments Print Email This Post

Silent Computer Upgrade - Part II

Posted by orion orion on Wed 13 of Jun, 2007 18:55 BST
In the last article, I talked about some of the research I had done to prepare for this. Some real thought had gone into it before I even started buying the CPU and considering the other tools and parts needed. Any project like this should have a little research done, although as this project demonstrates, you will also sometimes have think up quite a bit on the fly.

I first wanted to change the CPU, as I can change the power supply once everything else works. Unfortunately, I do not have any real means of measuring the wattage used by the setup, so I just estimated it. I am sure I could sit down and calculate it, but I had a 350 Watt supply at hand, and didn't want to buy another.

Tools

I now had the CPU and needed to make sure I had the other parts present. I had a basic computer toolkit, but most importantly needed these items:
  • Screwdrivers - at least 1 posi, and also one with a socket fitting for stiff screws
  • Anti Static Band - This should be clipped to an earthed item, even the computers case may be used as long as it is plugged in, but switched off. This is so you do not fry the computers parts with the natural static field surrounding your body. You may also get away with not using the band, and regularly grounding yourself by touching an earthed metal area.
  • A good, well lit area to work - don't I always recommend this? It really is important, for safety, for being able to work without being obstructed, to work with enough elbow room, and with good lighting so you can really see what you are doing. I really cannot stress enough that this is worth having for any tinkering with robots, lego or computers.
  • Tissues - there will be some heat transfer compound to wipe away, and I did a bit messy with it.
  • Mini Vacuum cleaner + cotton wool buds - The inside of a computer, especially around the CPU fan and heatsink will get very dusty very quickly. Make sure you have something to clean it a bit with.
  • Face Mask - if you are sensitive to dust, I would have one of these and wear it.

Safety Goggles are generally not necessary for simple computer modifications/changes, at least not unless the soldering iron comes out or you go beyond the basics. However, if you are working with heat transfer compound, do not rub your eyes, and make sure you wash your hands well before consuming any food.

Generally remember not to consume food or drink around a computer being worked on - coffee spilt on the motherboard would be a pretty nasty mess.

These are the tools I started with, but I soon found the need for others. More on that when I get there.

Parts

Some basic parts I prepared for this, although this is not exhaustive, as with the tools, more things became required as I progressed.

  • AMD Turion 64 ML-30 I bought last time
  • Heat Transfer Compound
  • Single fan quiet 350W Power Supply
  • Anti-Static? container to store removed Sempron CPU in.

Opening up the computer

This is actually very simple, and with my case - a Cooler Master Centurion, it was a matter of undoing one thumbscrew on the top, sliding the top panel off, and then lifting off the side panel. This case is a real joy to work with - it also has turned edges to save cuts and scrapes from sharp corners.

Just make sure you put whatever screws in a safe spot where you can find them later.

Removing the old CPU

The Sempron CPU already in the machine was, as mentioned earlier, getting pretty hot. I was using it with a genuine AMD Cooler and the heat transfer pad already present on it.

I unclipped the cooler, and normally a gentle tug should free them of the CPU/Bracket. However, it was stuck fast. The heat transfer compound was very thin, and had also hardened somewhat. The problem with Socket 754 and later CPU's is that the plastic bracket the cooler clips in does not allow access to the CPU free lever.

I had a choice of trying to run some software to really abuse the CPU and heat it up (like Prime 95, a suggestion on some websites), or to tug harder. I decided to gently tug harder. It is important here to make sure you are tugging completely vertically - any skew here will mess up the pins on the CPU, which is very bad news if you want to use it again. Luckily it came away safely (which considering I needed it again later was very important).

However, the CPU was now securely fastened to that cooler, and currently my plan was to use this cooler again. How could I get the CPU off the cooler? Well the HTC would clearly soften if there was a little heat present. I did not want too intense or focussed heat like a soldering iron. So I needed something more gentle with a large surface area. My choice was a travel iron on a low setting.

Since I am now applying heat, before I continued, I now made sure I had my goggles present.

  • Travel Iron - these are much smaller than normal irons, and have a few heat settings. Put it on a lower one - try to stay between 40 and 60 degrees. DO NOT put water in the steam making area. Steam and CPU's do not get along well together.
  • Safety Goggles - When heating (along with other things), there is a hazard of things spattering, and although HTC is not likely to, hot or cold, you really do not want that anywhere near your eyes.
  • Clamp - A gentle clamp, which I padded with some foam normally used to pack motherboards, to hold the cooler in place.

I put on the goggles, and clamped the cooler to the desk, then on a side of the cooler, I gently heated it up with the iron in one hand, and slowly slid off the Sempron with the other. I then cleaned the HTC from the heat spreader on the CPU, and put it into an anti-static bag.

Since I was going to use the heatsink and fan, I then cleaned it up. I used tissue to wipe away the HTC while it was still soft, and then a mini vacuum with cotton wool buds to wipe away the dust bunnies.

Fitting the Turion and heatsink

Now here was an issue that the Silent PC Review article had mentioned. The Turion has not got a heat spreader like the Sempron, and unlike the older open core AMD processors it did not have foam separators. I wanted to be sure that the CPU die would not be damaged by the cooler.

I initially though of pilfering the foam pads from a dormant Duron D600 I had, and use those as spacers. However, since that CPU was still last seen working, I preferred not to butcher it quite yet.

I then eyed up the cooler that accompanied this Duron, but could not find a good way to mechanically attach it to the motherboard, which is a shame because it was a smaller fan. I also considered using a fan/heatsink combo from a graphics card. These were put aside for what seemed a much simpler option...

My plan was to construct a paper gasket. This meant getting a pair of scissors, and some recycled paper - I often keep draft prints and similar paper so i can print on the back of it.
  • Scissors - any decent scissors will do.
  • Paper - Since you wont see it, it really does not matter what is printed on it. Perhaps one of those brochures that come through the door will do.

I measured the CPU package size first, and then measured the inner rectangle needed to clear the die and the small surface mount components around it. I cut about 10 copies of the larger rectangle, and then cut the smaller rectangle from them by folding groups of them in half and cutting into them.

I am not sure how many it took, but I put enough to clear the top of the core by a few millimetres, bearing mind that the paper will compress slightly as the pressure from the heatsink clamp comes into play.

I then placed the CPU into the socket, put the gasket around the die, and placed the heatsink onto it. It was quite clear that the heatsink was riding too high on the plastic bracket, and so I would need more paper to meet it. This also made me realise I would need a lot of HTC. This is not good.

About Heat Transfer Compound

HTC, also known as Thermal Grease is used to allow heat conductivity between metal surfaces, often used between a heatsink and a component needing the sink. It is not as good a conductor as the metal, and is generally used to ensure good thermal contact between two surfaces with no air bubbles or contamination.

It is often shipped with heatsinks, with many now opting for a pad directly on the base of the heatsink so only a protective sleeve needs to be removed instead of a tube.

HTC is not designed to bridge larger gaps, but more for surfaces which would be flush to each other but for the compound, possibly with some mechanical pressure.

In this instance, the gap was too much, and although I opted for the HTC - this is not a good solution. I only used it like that for about 3 days.

My advice to anyone else trying this is to purchase the ThermalTake? heatsink designed for use with a Turion processor like I later did, or to find an aluminium or copper plate of the right thickness and dimensions to fit in the inner rectangle of the gaskets, using a little thermal grease as designed above and below the plate.

Back to fitting the heatsink

I literally smothered the die in HTC, and used a scrap of card to spread it into the gasket like butter. I regretted that when I needed to clean it away later.

I then put the heatsink in, and felt the HTC being displaced as I wanted, then clipped the heatsink into place.

Powering Up

I powered up the board, and the motherboard recognised it straight away. However, going to the motherboard health settings, I noted that the CPU temperature was climbing to 40 + C pretty quickly.

I new that the thermal connection in the set up was not great, but I had another suspect - the Silent PC review mentioned that these processors were often overvolted by the motherboards, and sure enough it was. I changed the auto setting to the recommended 1.35v. The temperature began to drop nicely again.

I then booted the OS. The first thing I noticed was that the speed is reported at 800Mhz which is half of the CPU's actual potential, and that no stepping seems permitted. Unfortunately the board does not properly support PowerNow? on the Turion, so I shut down the OS and used the BIOS to change the multiplier. X8 was not good, although a CPU with an FSB of 200Mhz multiplied by 8 should be 1.6Ghz, the machine would power down after a minute. I then tried a few settings, and x7 with a somewhat increased FSB seemed to give me the correct 1.6Ghz.

The problem with the Turion processors, is instead of the Desktop CPU's which use software drivers to clock down when there is little demand from a default state of full speed, the Turions will clock up when power is needed from a default state of a much lower speed.

The BIOS did not show any setting to enable Cool N' Quiet or PowerNow? technology that can be used to control an AMD CPU's speed with software. I booted Ubuntu, and tried to load the powernow-k8 module but received the message "Device not found". Checking the DMESG log there were a number of messages to the effect of this feature not being supported by my BIOS. Net result - no software controlled CPU stepping.

However, the OS was now running fine, and stable. I did feel I was not getting the full potential to save power, but the temperature was very good, at between 32 and 35 degrees C.

In an attempt to get BIOS support for this, I went to the Gigabyte (motherboard manufacturer) support site and downloaded the latest flash code. Now - I do not have any floppy drives, so I needed to make a bootable CD.

Making the bootable flash CD

I used Ubuntu Linux for this. I unpacked the file downloaded from Gigabyte, and also downloaded a Freedos boot disk image to tailor to my needs. I used a boot disk known as "Ripcord".

With this boot disk, I was able to mount it as a loopback vfat device, and copy the files from the gigabyte archive into it.

mkdir /mnt/bootdisk
mount /home/danny/freedos.img /mnt/bootdisk -t msdos -o loop


You will need to run those commands as root, or by prefixing them with sudo as is the convention in Ubuntu.

The files to copy from the gigabyte archive consisted of an autoexec.bat, a flash utility and the image itself. I then built a bootable CD using the disk image.

I booted off the CD, flashed the BIOS and rebooted....

Disaster

The motherboard would no longer boot. It would go through the graphics cards own boot screen, then immediately power down. No beep codes or error messages, just power down.

I had a sneaking suspicion that the new update (F19 - with a change log message of "CPU Codes") might no longer support the Turion. It is a good job I kept the Sempron around, as I planned to try booting with that and reverting to an earlier F18 BIOS image the next day. I will tell you about that in the next part...


WARNING: No such module TAGS! PC Upgrade Silent AMD CPU Turion Sempron Ubuntu Linux Bootdisk Gigabyte Motherboard FREEDOS

Permalink 0 comments Print Email This Post

Silent Computer Upgrade

Posted by orion orion on Sun 13 of May, 2007 14:50 BST
As well as building robots, I also like to build and upgrade my computers. In fact, I like generally tinkering with all sorts of machines. My main desktop was once specified and built as a hard core gaming rig. It was a little past that now. But it had a bit of a wopping specification, with all the noise and power usage problems that come with it.

My Specs

  • Case: Cooler Master Centurion, Aluminium Case with blue power and HDD LED's. Slick, and cool.
  • Geforce 6600 AGP Graphics Card - With extra power connection and large fan/heatsink rig.
  • AMD Sempron 2800+ without CnQ - this had a fairly loud fan, and ran between 45C and 60C. It got hot enough that I sometimes worried. It has a TDP of 62W according to Wikipedia (April 2007).
  • 1Gb Kingston DDR Ram.
  • 1x80Gb (Windows XP) + 1x160Gb (Ubuntu Feisty) IDE Hard Disk Drives - Noisy, power hungry and hot.
  • Gigabyte Triton K8NS (Not the Pro) motherboard, with NForce Chipset - including on board AC-97 sound, Ethernet etc.
  • Belkin Wireless-N 802.11(a/b/g/n) card.
  • Internal DVD+/-RW Drive.
  • 3 Additional Case fans to keep this cool.
  • 600W Gold Plated Triple Fan Power Supply.

I suspect you begin to imagine what kind of monster this was. It ran hot, it hogged power, and it sounded like an jet taking off when started up.

It was time for an upgrade. Now most applications I run, run smoothly enough on this rig, for me not to be too worried yet about going Duo Core or some other next generation architecture (at least for now), but I had a goal to reduce power usage, noise and heat generated by this machine. To turn it from a hard core rig, to a super silent (or as close as can) kind of machine. Considering how different these use cases are, this could be quite interesting.

First Steps

Some simple things got the first few parts out of the way. I replaced the Internal DVD+/-RW with an External USB-2 one, which I can turn off when it is not needed.

The 80Gb hard disk was also moved to an external caddy, it is the Windows drive, which I generally only use for gaming. Again, this is USB-2, and I am a little worried about game load time.

Until I make some other savings, I cannot yet replace the power supply. I still have three case fans, the Sempron, its fan and the 6600.

The CPU

The CPU is the most interesting part of this upgrade. I read an article on about putting an AMD Turion 64 Laptop processor on a desktop motherboard. The interesting thing is that the Turion 64 uses the same socket 754 technology as the Sempron, and I might be able to fit it.

This would bring some great advantages. Not only was it a faster rated processor in terms of performance (same clock speed), but it was 64 bit. More importantly, it had a TDP (Thermal Output- Typical Dissipated Power) of around 31W - half that of the Sempron. This meant it used less power from the PSU and needed less cooling (lower fan speed or simply a smaller fan). It was obvious this was a good route to my goals.

I started doing the research - would my motherboard support it? I did not find much information specifically regarding my board, only one article on using the Turion with the K8NS-Pro motherboard - one that had a number of differences from the non pro model. The research suggested I might need a different BIOS image depending on the age of it and what kind of stepping it supported. With a Sempron that did not support Cool and Quiet voltage stepping, I had no way of knowing. Gigabyte suggested it might not.

I decided to take the risk, and buy one of these. I had to hunt around and shop around, and eventually found one on eBay for £65. This was the AMD Turion 64 with a Lancaster Core- the ML-30. I was actually able to collect it locally, so I was able to get underway within a day of starting. The CPU was still in a sealed static proof bag, so it was in new condition.

My next step was to remove the old CPU and fit the new one. This threw up a few issues, so I will cover this next time...


WARNING: No such module TAGS! PC Upgrade Silent AMD CPU Turion Sempron Gigabyte motherboard

Permalink 0 comments Print Email This Post

OrionRobots and I pledge our support...

Posted by orion orion on Thu 03 of May, 2007 19:34 BST
..to the campaign to free the reporter Alan Johnston. It is appalling that such a man has been in this situation for such a length of time, and no group holding him are deserving of any concessions or respect for such an act.

We salute you Alan, and hope, along with many others around the world, that you are returned alive and well.



Permalink 0 comments Print Email This Post

Interesting rodent behaviour

Posted by orion orion on Mon 09 of Apr, 2007 19:12 BST
The animals around us are clearly, much like us, and the robots we aspire to build, some collection of complex sensors, actuators and behaviours. These form the inspiration for our building, and sometimes certain recent science reveals surprising things that may help in building robots with such behaviours. As such, I am always interested in these behaviours, as well as animals with extra-ordinary senses or limbs. One of the great things about robotics is that you get to be multi-disciplinary like that.

Had Da Vinci not thought about animal anatomy and behaviour, perhaps he would not have designed his flying machines, tanks and robot (seriously - this is not a joke, read Leonardo Da Vincis Robot ).

More recent uses must be the BEAM technology used by Mark Tilden creator of the Wow-wee Robosapien.

Anyway, I recently spotted a couple of recent studies on animal behaviour which were most definitely interesting, and may also push their estimated IQ's up a notch or two. I will probably post other articles on them later.

Rats Cognitive ability better than once thought

Rats showed a few interesting abilities in a recent study at The University of Georgia, published in the Current Biology journal earlier this year.

The study, written up as a paper by Jonathon Crystal and Allison Foote, involved sound tones, with a fuzzy definition of "short" or "long", and a method of recording a decision. The rats who chose the correct duration were given a food reward, those who chose the wrong answer got no reward. A third alternative (not always available)- of backing out of the test altogether gave a much smaller food reward. This is known as a "duration-discrimination" test.

So the rats could gamble this smaller food reward for a potentially larger one. This shows reflection and retention of knowledge. Rats seemed to remember how this worked after a few tests, and where allowed, leave the harder ones for the smaller food rewards, taking the easy ones with extreme values of long and short. Also, over the whole test set the rats seemed to do worse on the tests without the alternative third option.

The results showed an overall correlation that rats may have actually understood when to back out and when to gamble. It shows an ability to think about thinking, to understand the "I just don't know" state. This kind of ability could be referred to as metacognition.

One big question is if this is concious behaviour. It is certainly learning and use of acquired knowledge.


WARNING: No such module TAGS! Robots rats metacognition UGA cognition reflection learning

Permalink 0 comments Print Email This Post

NXT - Building the Arm

Posted by orion orion on Sun 08 of Apr, 2007 22:48 BST
I have continued my exploration of the NXT kit, and the instructions within. This part gets interesting with a robot arm.

Disassembling the Tribot

The first task is to disassemble the Tribot - which takes a bit longer than the studded models, and takes a fair bit more dexterity.

Axles and pins can start to be annoying when pulling many out of a beam, and I have heard tips about using marigolds or a little strip of rubber to grip them and save your fingers.

For the axles, the Flexible Axle part discussed on my blog Saturday 11th Jan can be used - note they are all currently used in the ball holder. You should also use the push through technique with another axle. I am not sure about marigolds - they will limit my movement, and may make it difficult - but there's time to give them a go.

I start by removing the NXT itself, and the cables. I got stuck in, and as it turned out, the only real tool I needed was an Axle to push out pins and axles - this took me only a few minutes.

Building the RoboArm T-56


Now this is where the build starts to get really interesting, and my use of studless techniques will begin to be tested.

I got stuck into the arm. This took a little longer, and initially, it is not much like an arm. But it has enough to rotate into position and touch things, along with two hand cranks on the sides to manually override it. The arm itself cannot be positioned without the cranks because it uses a worm gear based drive. This also means that it is pretty slow, even on full power. However, it is also steady.

Programming it


I tried the simple initial program. I had to reduce the given values a little, as my bot tended to push down hard enough on the touch to either bowl the ball off or lift the arms base. A reduction to 9 rotations on the initial downward position did the trick.

Building the hand

Well the arm is not much fun without a hand. A hand will be useful to pick up a ball and move it. Jumping ahead - I suspected that to be the challenge.

It was an interesting build - the use of the knob wheel again. I really have not used these all that much at all. On the second set of claws, I actually put the ball pin in before the knob wheels - as it looked like it would be a better plan.

Also, if it looks like this part of the arm is not quite coming together, have a bit of faith and carry on -it does work. Wire routing is much easier here than for the buggy, but for the stuff going under the NXT I had to remove the clips on one side, to lift it up a little to get the RJ plug through.

Programming the hand


This was a more interesting program - the switch block (or If block - as it is only really if/else and not multi-case) was introduced.

Again, after trying it out in a vanilla version, I had to tweak values when testing to make sure it did not overshoot. It may have been more clever to use a wait until touched for the downward movement, but since OR blocks have not been introduced, it could have got messy when it couldn't go any further.

Even with much tweaking, it either grabs the pedestal with the ball. knocks the pedestal over before getting the ball, or knocks it over when it gets the ball. It also tends to knock over the other pedestal when dropping the ball. But at least it did pick the ball up and move it. I managed to reorganise the program so the arm lifted a little after machine the grab, and then moved, thus not knocking the pedestals over when it did so.

Fitting the light sensor

This will be used for sensing colour. It is a small build, but manoeuvring in the sensor assembly is a little tricky.
I also reduced the pedestal height from a 12 axle to a 10, which helps to get the arm over the pedestal without knocking it over.

In programming it, I again started with all the recommended values, but had to tweak the light sensor to the right threshold to stop them all coming out as "blue". The sound (speech) was pretty meek, so I upped it to 100%. And even with the reduction in height of the pedestal, the grab still seems to knock the pedestal over most times. Oh well.

The map tab (for the extents of the NXT G-Code) is beginning to show its limits, and some normal scroll bars would be welcome here. My laptops touchpad can be configured to scroll horizontally with such bars.

After all the tweaking, the identification works, and it will move the red ball to another location as expected, if still occasionally knocking over the pedestals.

What is next


Well - in the NXT guide, we have animals next - the scorpion - Spike.. Cool..

Until NXT time then!


WARNING: No such module TAGS! robots robotics lego mindstorms NXT

Permalink 0 comments Print Email This Post

Stuff For Sale

Posted by orion orion on Sun 08 of Apr, 2007 19:26 BST
I am selling a couple of bits on eBay at the moment, and thought I would shamelessly plug them on my Blog.

There is a gold plated 650-Watt Triple Fan power supply. It is a monster, and pretty cool at that. However, with my current focus on making a very cool, quiet, low power machine, the 650-Watt serious gamer power supply did not really fit any more. It is still relatively quiet, but not as quiet as the silent technology that I have spent a whole week fitting my computer with (writeups coming this week!).

Q-Tec 650 Watt Triple Fan Gold Plated PSU (external link)

Also currently being sold is a 5 port KVM switch. I have been scaling back my little farm of boxes at home, in favour of using VM's to achieve the same result as 4 completely underutilised machines. It is one of the solid ones with a real rotary switch, and it was bought new in 2003.

4 Port KVM Switchbox (external link)

I may be selling more stuff, as well as freecycling the stuff I think has lesser value. I will keep my readers posted.

View all my items on ebay (external link)

WARNING: No such module TAGS! Hardware PC PSU ATX KVM

Permalink 0 comments Print Email This Post

A Grand Challenge comes to the UK!

Posted by orion orion on Wed 21 of Mar, 2007 23:40 GMT
The UK Ministry Of Defence (MoD) have decided to take a leaf from the DARPA book, and created their own grand challenge.

As the original did, this will intrigue and bring forth robot builders, makers and inventors from all over the country and potentially the globe to take part. I will certainly be keeping an eye on this.

So what is the challenge?

The MoD would like to encourage design of an autonomous vehicle capable of scouting out urban areas, and assessing for any potential threats. The vehicle must be able to remain undetected where possible. It should be able to deal well with returned fire if discovered, or be cheap enough to be expendable.

This is already enough for a serious challenge just to make an autonomous vehicle that can travel any reasonable distance in terrain that is not necessarily well mapped or known. Experience from the DARPA Grand Challenge so far has shown how difficult this can be.

To then actually gather intelligence and report back ads a further element, where sensory and telemetry equipment must be thought through. The initial draft of applicants to the challenge, opened by the Defence Minister Lord Drayson, may only get this far.

Then the other important part is to remain undetected. This means low noise drive systems - which could be pretty difficult on rough terrain like rubble. A good acoustic profile could mean everything when trying to avoid detection. Since it may not be easy to get a robot in position back to base, then transmission protocols must be further refined to ensure that they are not easily detected or intercepted, and long range enough to allow the intelligence to get back to the troops. The design will need to have thought out profile - keeping a small, or inconspicuous shape, as well as camouflage of sorts, so it does not stand out in the environment.

In short, it would have to be designed taking into account some of the training that those who would be involved in stealth operations would have had.

Keeping costs down is played off versus actual armour or ability to take fire. Of course excellent ability to avoid detection may mean neither are necessary. Stealth is by far the most important factor, as it would also alert enemy troops that friendly troops are moving in.

Where, how and when?

The competition is open to all kinds of companies, institutions and academic groups, with the MOD proposing to finance and offer a development contract to the winners. When putting in the R & D for this, competitors may apply for funding in the form of grants from the MOD, but depending on their status, they may also be required to fund themselves.

The first event dates have not been announced, only that things will be kicked off (possibly for team applicants only - no spectators) at Copehill Down in Wiltshire, UK around January 2008. The grand finale will take place late in the Summer of 2008.

You have only until the 15th May to get your applications in - you can enter through the website, and will go through a qualification process.



Permalink 0 comments Print Email This Post
First Page Fast Prev Prev PagePage: 5/19 Next Page Fast Next Last Page
1 3 4 5 6 7 19

Current Poll

Are you interested in a London robot building Group?
Yes I am interested in a senior group
Yes I am interested in a junior group
Yes I am interested in mixed group
Already part of one (please leave the details below)
Do not live in greater London
Not really my thing
Other (please leave a comment)

View Results
(Votes: 11)

Online users

13 online users
RSS feed Wiki RSS feed Blogs RSS feed Articles RSS feed Image Galleries RSS feed File Galleries RSS feed Forums
Babelfish (debug): Fatal error
[ Execution time: 0.47 secs ]   [ Memory usage: 19.30MB ]   [ GZIP Enabled ]   [ Server load: 10.51 ]