Arduino Switch Box

arduino_switch

When you’re prototyping a new project, sometimes all you need is a switch. The folks at oomlout were tired of constantly having to rewire things, so they built a universal switch box for the Arduino. It has five potentiometers plus three switches. They’ve put together a software package that monitors the switches and can show you a live view of the knob positions. Have a look at the video below for a demo.

The writeup actually hints at what we can only assume is the next kit they’re releasing: a robot arm. Continue reading “Arduino Switch Box”

Xbox 360 Portable

ben_heck

A couple months ago we posted [Ben Heck]’s in-progress photos of his Xbox 360 laptop (with links to his other versions). He’s just put the finishing touches on it, and dubbed it the Xbox 360 Portable. It has a removable hard drive on top and memory slots on the side. The webcam is embedded in the frame and there’s internal WiFi. With chatpads available now, he’s decided not to include a keyboard. It’s really a nice machine. Check out the video below for a tour of the system.

Continue reading “Xbox 360 Portable”

Wiimote Finger Tracking Music Controller

[youtube=http://www.youtube.com/watch?v=Y7uLuYG62sY&fmt=18]

The Evolution Control Committee has been doing live mashup performances for many years and recently upgraded their hardware. Inspired by [Johnny Lee]’s Wiimote whiteboard, they built a rear projection display they could use during performances. It displays a dense collection of samples in Ableton Live. On each of the performer’s hands is an IR LED mounted to a thimble. By touching the thumb to the forefinger, the LED turns on. Two Wiimotes watch for these IR flashes to trigger mouse clicks. [TradeMark G] found the Ableton display too complex to navigate quickly and accurately with a mouse; this new display make things much easier and enjoyable.

[via Laughing Squid]

G1 Multitouch Proof Of Concept

[youtube=http://www.youtube.com/watch?v=pSBYqmWVqeM]

[RyeBrye] has been trying to get multitouch working on the Android based T-Mobile G1. He hacked the Synaptics touchscreen driver so that it would dump raw event info to a character device. The demo above is using example code from Google for a fingerpaint program. Polling the device is not the fastest method, but [RyeBrye] just wanted to get a demo out there to prove it could be done.

CuBear, Berkeley’s Rubik’s Cube Solver

[youtube=http://www.youtube.com/watch?v=N7BksZZyAdc&fmt=18]

A team of five UC Berkeley engineering built this impressive Rubik’s Cube solver. The CuBear is a giant transparent cube with a servo attached to each face to rotate the cube’s six faces. The user can either scramble the cube using computer controls or show the faces of a scrambled cube to the onboard webcam, and the machine will replicate it. While scrambling the cube may take many moves, the computer calculates the shortest number of moves to solve the cube before proceeding. Team member [Dan Dzoan] is quite a fast solver himself, as you can see at the end of BotJunkie’s video embedded below. Continue reading “CuBear, Berkeley’s Rubik’s Cube Solver”

Stupidly Huge POV Display

pov_gear

[Mario Mauerer] and friends were commissioned to build this 2m persistence of vision display (translated) for a party (in a hight-voltage lab no less). Dubbed “Display from Hell”, it uses 100 blue SMD LEDs to generate the POV images. They’re connected to an ATMega64 via shift registers. Their target speed is 600rpm for a flicker free image, which means the propeller tips are moving at 140mph. The board can be updated wirelessly via IRDA and plans for adding SD storage are in place. You have to see and HEAR this thing in the video embedded below. Continue reading “Stupidly Huge POV Display”

Tennis Ball Fetcher

tennisball

Reader [Julian von Mendel] and his team built this tennis ball fetching robot for a competition (translated). The first version used distance sensors to locate the tennis balls for pick-up, but they changed to a camera based approach. The robot has three omniwheels and is designed to calculate the shortest path to the ball despite orientation since it can rotate while traveling. The wheels are monitored using rotation sensors from PS/2 mice. The control is provided by 3 Atmel microcontrollers that communicate via SPI. The multiprocessor design is fairly generic and could be reused for a different style of robot. While their robot performed fairly well, there were some shortcomings. The limited storage space meant frequent trips to drop off balls. The tilting bucket kept them from picking up tennis balls that were against the wall. Also, the bot had to be disassembled for battery swaps. The project is very well documented and they’ve released all of their control code. You can see the robot retrieving a ball after the break. Continue reading “Tennis Ball Fetcher”