Get an mbed!
Subscribe to the mailing list
Scott O'Brien has blogged about an underwater robot he has built for his final year individual project at the University of Westminster using a couple mbed microcontrollers.
The craft consists of a PVC frame and uses seven modified bilge pump motors to level and move the vehicle.
As well as controlling the motors, the LPC1768 mbed in the bottom controller interfaces various sensors including an IMU board fitted with a 3-axis accelerometer, a 3-axis gyroscope, a 3-axis magnetometer, and a micro-SD card socket.
The top controller features an LCD display with two analog thumb joysticks to manoeuvre and uses RS232 to interface the ROV.
Read Scott's blog here
The latest mbed compiler is now available in beta!
This release primarily has behind the scenes changes and bugfixes, but there's one improvement you should definitely notice: speed!
We have added incremental compilation support, which means that when compiling large programs, subsequent recompiles will be much faster, only recompiling those files which have changed. There's also performance improvements to the UI in several areas.
Try it out at betamode and please let us know what you think and report any bugs in the usual place. We'll aim to roll out the new version to all users in the next week or two.
This is an awesome mbed project using Google Calendar to remotely control a home heating system.
mbed developer, Shack Dweller, describes using an Ethernet-connected mbed NXP LPC1768 Microcontroller, a DS18S20 temperature sensor, and a relay to control his heating unit based on events setup in a Google Calendar.
He describes the Google API developers guide as "scary read at first" but simplifies the process by using private address for his calendar and avoiding authentication. Using Google API GET requests, he polls the calendar to return events in XML and uses the current ambient temperature to switch the heater using a solid state relay. He has posted his code here
Great project with countless opportunities, great stuff.
Read more about the project on his blog
A team of innovative engineers have been working on a haptic-feedback device to assist the visually impaired to navigate around obstacles using an MS Kinect and Panda board, for machine vision and image processing and an mbed for driving the haptic-feedback.
Jatin Sharma, 
Tushar Chugh and Rolly Seth have written a blog about their project and the challenges they have faced.
Tushar Chugh wrote: About viSparsh
"viSparsh" is a fusion of two words, " vision" and "sparsh"(touch), which depicts the purpose of the project to aid visually impaired people through the sense of touch. A team of three Young India Fellows Jatin Sharma, Tushar Chugh and Rolly Seth are developing this haptic belt under the tutelage and mentorship of Prof. Rahul Magharam, Director mLAB, University of Pennsylvania’s School of Engineering and Applied Science.
 |  |
What is the focus of the project?
The focus of the viSparsh project is to create a haptic (touch-based feedback) belt for the blind with a simple goal that a blind person can avoid obstacles that are on and above ground height. This belt has an infrared optic sensor that projects thousands of dots in the 3-12 feet distance to determine what obstacles are present. Based on the proximity and relative direction of the obstacle to the wearer, a series of vibrators on the belt alert the wearer of the presence of obstacles around them. This system is lightweight, low-cost (Rs 5000), easy to use, non-conspicuous and also frees up the ears and hands of the blind person. The system provides a 160 degree view of the environment and is complementary to the limited reach of the walking stick. It also does not require any additional infrastructure in the environment.
What kind of field work was/is done?
We have evaluated the belt in different work environments. The first type of evaluation is in a room with obstacles placed at fixed positions. This allowed us to measure the obstacle tracking error, the guidance algorithms in the presence of multiple obstacles and also obstacles of different sizes and at different heights. We have evaluated the usability, battery life and overall response with multiple blind persons. The second type of tests is in a "real-world" scenario with the wearer walking through a busy corridor and a busy room with many people. The response from the blind persons has been very positive both with respect to the wearability and the natural touch-based feedback. We plan to deploy more units with more compact on-board electronics in the near future.
 |  |  |
Problems Faced:
What kind of an impact can the haptic belt make for people who are visually impaired?
The everyday life of a blind person is often taken for granted and the only navigation help is the walking stick. The stick, while effective, has limited reach and often does not detect obstacles above ground. Our focus in this project is to make a technology-packed device that is affordable to the average Indian and easy to use with just 5 minutes of training. While our focus is on the technological aspects, we strongly believe that technology is a means and must be complemented with user-based feedback with extensive evaluation and field tests. Our team logs every sentiment the blind person wearing the haptic belt gives us and we take their feedback very seriously. Our eventual goal is to deploy several hundred of these belts to everyday users so we can make a positive impact one person at a time.
More info please visit our blog: http://www.visparsh.blogspot.com/
Well done guys and thank you very much of the guest blog. Great to see an mbed used in such a inventive project, we hope it has helped you prototype your design concept, and hope that your belt eventually helps blind people all over the world!
