AlmightyArjen

The rubber-poking idea is indeed a good one! But the movement is a bit difficult: you need to poke and pull at the same time in a reach of just centimeters. I have thought about a second device to push it open from the inside, but the whole idea is to make a robot that can do everything by itself. I'm now building the grabbing arm that actually picks the a can from the fridge. I think I can use the "fingers" to open the fridge further from the side...

Looks really nice indeed! Lego spends much time/money on a good website; they know the importance of it!

Looks enormously like the original and it looks really awesome! Nice job!

At first I need the robot to approach from the front to open it a little bit. Since there's a considerable amount of force needed to open the door the first centimeter, I cannot do it from the side. After that, I can leave the fridge door open (when I let it go when it's open for about 10 centimeters, it closes not completely again but leaves a gap of a few centimeters) and drive to the side to put in some kind of long pin and then drive sideward to open it further. The robot is indeed noisy but I want it to be all Lego so I don't want to change the motors to different ones :) The grabbing arm will have a colour sensor. By scanning the colour of the can/label you can have different choices like beer (i.e. green cans like Heineken), coke (coca cola red/Pepsi blue) and so on...

Very nice, I like the design! It's also cool that you've managed to put in a working V6 while the model is quite small!

You all get a beer when it's finished ;) There is an Arduino shield for NXT motors available at mindsensors: http://www.mindsensors.com/index.php?module=pagemaster&PAGE_user_op=view_page&PAGE_id=185 I've ordered one and it works really great. You can connect up to 4 NXT motors and a bunch of NXT sensors (I doubt about the use of the NXT sensors, as one has already decided to go for the Arduino platform but okay...).

I came up with the idea to build a robot that can get me a beer or soda from the fridge while I'm watching Die Hard... I'm building it completely from Lego Technic, mostly studded. I use my own designed electronics to control the motors, pneumatic valves and read out the sensors. It saves me hundreds of Euro's by not using the NXT platform ;) Next to that: the NXT platform is not as flexible as other programming platforms. I have been busy with it for a few months now and there were some problems to be solved so I've divided "the project" in smaller goals: 1. Get the rover over a bump in the living room (success!) There's a nasty bump between the kitchen and the living room. Since the robot is going to be well over 10kg, I had to find a way to get it over the bump. The solution was to use caterpillar tracks (the robot needs to be able to turn around its own axle) with on top of that studless beams for extra grip. I've put some filt on top of the beams so my wooden floor doesn't get damaged. Video 1: 2. Make the robot move sideward to be able to position itself (success!) Not being sure about the precision of the tracks, I need the robot to be able to move sideward for positioning itself in front of the fridge (to be able to grap the handle). I designed a pneumatic system with 8 pistons that push down 8 wheels, that lift up the robot just a bit. That way the robot can move sideward, which is impossible with caterpillar tracks. As you will see in video 3, the accuracy of the distance sensors is good enough that the robot doesn't need to move sideward. If the system isn't needed when the robot is completed, I could remove it. Video 2: 3. Indoor navigation: make the robot find its way to the frigde, couch and back (success!) Using 4 ultrasonic distance sensors and a digital compass, I make the robot navigate around the house. It works just perfect! Video 3: 4. Think of a way to make the robot open the fridge I'm working on this one now. This one is by far the hardest to complete. I have already a pneumatic system that can grab the handle and open the fridge a few centimeters by setteling itself against the fridge. As you know: sometimes the fridge door is a bit stuck and you need some force to open it. Grabbing the handle and just drive backward doesn't work because of the height of the robot. Right now I need to find a way to open the fridge door further and move the robot to the inside side of the door. This is tricky and I haven't figured it out yet. If you have any suggestions: let me know! Further things to do: 5. Make the robot able to grab a can of beer/soda 6. Make the robot do the whole thing: getting out of hibernation, getting a beer, bring it to the couch and go back to the hibernation spot 7. Design an Android app that communicates with the robot through Bluetooth so I can order a beer from the couch

Yes, I use Arduino: it's a great platform with loads of libraries available. And since I'm a bit lazy: ideal... :)

Hi everyone! Just saw this "introduce yourself topic" so here I am. I'm lurking around for a few months now and so now and then I post my own creations. I discovered that this is a really cool forum with loads of people who are even more enthousiastic about Lego than I am. For me, it all started in 1985 when I got the 7727 train for my birthday. I had never been so happy! Since then I got some more Lego sets and when I grew older I switched to Technic and bought the classic 8868 and the 8880 for myself. Now I combine Lego with electronics: I'm building all kinds of robots and recently I've built a huge automated train track (see my posts). I don't use the NXT platform; it's quite expensive and I like to develop all the electronics myself. Since a few months I'm building a robot that's going to fetch me a beer from the fridge but there are some obstacles to take. I'll open a topic about that soon. Cheers and let's build cool things!

Thanks for all the nice comments. I already designed the track before I built it so it took me only half an hour to build the whole track. Laying and connecting all the cables took me about 3 hours. After that I needed to optimize the programs (which I had already written): when the train passes the IR sensor, it takes some time before the track can be powered down when the train is at the station, that kind of stuff. This optimizing took me also 2 hours I guess. The shot from the train: it was nice indeed but the wagon+camera was topheavy and kept on falling in the curves. I had built a special low wagon for the camera as well but this had the same problem. I could have adjusted the speed of the train by reprogramming the controller but I didn't feel like to do so. Now I regret it ;)

And as promised: the how-it works-video :)

Very cool!

I used IRZ48 power MOSFET's to switch on/off the tracks. It works quite simple. These FET's are my favorite: the can deliver up to 50 amps so there's never a shortage of current :) I've used 4 Sharp distance measurements to locate the trains. There are also other questions in another forum: I'll make a "how it works video", probably this afternoon and you'll know all there's to know ;)

One reason: girlfriend

I'll make an explanation video about it, there are more questions about it. I'll let you know. It's quite simple actually.

I've built a huge Lego train layout which is fully automated by microcontrollers, Arduino. The passenger trains 7740 and 7745 ride around and after a while they stop at the station. The freight trains 7755 and 7760 intersect with the track of the 7745. A section of the track of the freight trains can be powered down: if the 7745 passes the intersection, the 7755 and the 7760 will stop. The trains use PWM signals to start and stop. For starting is actually needed: otherwise the locomotive will disconnect from the wagons and drive away by itself. Video: I've built it in my living room so unfortunately I have to tear it down again :(

Indeed, the extra weight causes more friction. If you take off the weight (or take a original 9V train like the metroliner), the train has much more wheelspin when turning on the power suddenly. Disadvantage is the fact that slopes are harder to climb: the weight/friction relation is not linear: more weight doesn't mean more fiction and thus pulling power when going up :)

Very cool! I like how you kept the colour scheme exactly the same as the Space Police sets. Nice job!

The 7740 uses the RC wheels since it had to pull lots of wagons. The 7727, which looks quite fast in the video has its own old style wheels... I had to know so I just ran a few tests: 1. the electrical resistance doesn't change if you use a full weight 12V locomotive or just the motor itself. When I pressed really hard on the motor, I could make a difference of 0.5 ohms, which is negligible. 2. the regulator output is a nice and stable 9,1 Volt 3. I've rebuilt a track with the same straight length. I first measured the time of one straight length (16 track sections) in the video with a stopwatch, which resulted in 2.1 seconds. Then I measured the time of my test track, which resulted in 1.9 seconds. The difference can be explained by the fact that the straight track in the video is more distant from the voltage regulator connection that it is in my test track. Conclusion: the trains just go THAT fast... Looking at it, I had to turn down the speed of the cargo trains a bit while shooting the video. Good lesson for my next video

That's exactly what I thought! I checked everything: also the video settings of my camera and the video editing program and everything is as normal. While shooting the video I was already thinking that the trains were running quite fast. For the 7755 with the two motors I had to turn down the transformer a notch, otherwise it just derailed in the curves. So assuming that the video is correct, why do the trains look disproportionately fast? Is it because the 12V motors don't run that fast and we are not used to see "12 volt trains" move so fast? Or is it because I left the heavy weights in the locomotives, what causes a higher pressure on the rails and so the electrical resistance between the rails and the wheels is smaller?

I've finally decided to go definitely for the 9V tracks. I've bought a bunch and the trains run really smooth! The trains have less pulling power so for some locomotives it's necessary to use two motors, like on the 7755, what makes it a real work horse! I've made a video of my trains running on a 9V track:

That's indeed a good option antp!

A while ago I got my first 9V set: 4555 cargo station. And that was the first time I got to hold 9V tracks and the tracks looked really solid to me. I have a lot of 12V tracks laying around and 6 or so trains. The problem with the 12V tracks is that they're getting old and taking it apart results in broken click systems (the little pins on the sleeper plates). I was thinking to keep a short track and two switches for nostalgic reasons. I don't have many remote controlled switches or other "extra stuff" that isn't available in the 9V system. Next to that: I'm handy with electronics so I can fix it myself if needed. Further: I don't have a fixed layout laying around: I don't have space for that. I'm choosing the old 9V system since I want it to be powered by a transformer: I hate batteries... Pros: - 9V more solid system, taking apart is easy - trains run more smoothly - I can sell the 12V tracks (the ones that are not broken) pretty easy on Bricklink (I want to keep the 12V motors though) and buy 9V tracks pretty easy too. Cons: - It's not all 12V anymore: maybe it loses it's charme - No easy remote controlled switches and other extra stuff What do you think about making a switch to keep my trains running for at least another 20 years?

Thanks. Nice video: never thought it would work like that! But I think the shape is too sophisticated to build it with lego bricks...

Hi! Didn't know this forum, pretty awesome! I want to show you my technic robots that I've built so far. Since I'm from the early 80's, I never got used to building with studless (although I see some advantages) so forgive me: the robots are all studded ;) Next to that: I love electronics and I love to design it myself so I'm using the Arduino platform instead of NXT to control my robots. For me it's more flexible (and cheaper). Plant watering robot This was an idea of a colleaque of mine when I was telling that I was building an autonomous rover (see below). He said to me: why don't you build something useful like a robot that takes care of the plants while you're on holiday! And an idea was born... See below the result :) Autonomous rover which returns to a charging station when battery runs low This one was a real hassle to get it work. The rover itself wasn't the problem, but to make it park into the charging station was tricky. But I pulled it off by a coded IR signal which tells the rover when to turn and drive into the station. It's built in a Classic Space theme with cool lights. Oh, and in the end of the video, the rover takes care of some evil space pirates ;) Light chasing autonomous rover Predecessor of the rover above: driven by NXT motors it was a "real rover" with wheels. Unfortunately it wasn't agile enough to make it park into a charging station...