Blog - Sketching with Hardware

Day 5

Published on: | Author: Sketching With Hardware | Categories: 2014b, Daily Logs

At the fith day, the project status of the teams are very different. Some have already finished their housings or are working on a very detailed level, others are still challenged with big mechanical or electronical problems. Although there are long waiting periods to use the laser cutter, most of the teams are very optimistic to hit the deadline at tuesday (perhaps with some extra hours at the weekend!).


Group 1


With the aim of hacking their vehicle to an annoying robot with the name „the great Carholio“ (inspired by Beavis and Butt-Head’s character „the great Cornholio“) Team 1 had to work with two type of sensors. One to identify and follow the user and one to check the directions towards the user. So, for the orientation of the vehicle towards the followed person, they chose an infrared sensor, which is usable for short distances, without any sources of irritation. This already worked perfectly fine, using a remote control as a sender for the infrared signal. For the bigger distances, they worked with a bluetooth module, which caused some challenges, controlling it by the AT-commands.

In the beginning of the project, their idea was to work with a camera shield – taking photos and transfer it to the internet as a kind of live blogging bot. But the delivering time for the Arduino camera shield was to long for the project term – so they reduced their functionality to a sound module, which produces annoying sounds according to the speed and the direction of the user towards the object.

Also the model building of the housing was going well and the team was very optimistic about the next steps and their time management.


Group 2


After finishing their first rapid prototype of their „dealing mechanism“ which consists of a hacked servo and a rubber wheel of an old laser printer, they spent a lot of effort to construct a nice laser-file to rebuild the housing in a more precise way. In this prototype they also thought about the ergonomics, to have a very easy handling to adjust the robot and to lay in the play cards.

Also the software is running on a high level. Therefore, they wirk with two distance sensors. One at the back side and one at the left side – the Arduino identifies the end of the table and starts the navigation around the edge. With a third sensor, the vehicle is able to detect if a person is sitting in front of it and gives the signal to drop a card. The whole method works already, but a lot of effort is needed for the detail adjustments. Also half charged batteries caused big problems to the final adjustments and also some danger for the vehicle to fall off the table.


Group 3


With the idea of building a „sneaky“ vandal vehicle, masked as a „Alpina Weiß“ paint pot, group 3 built a servo controlled mechanism to regulate a built in spray can. This already worked very percise and reliable. The robot should then be able to follow a programmed path and mark this with the paint out of the spray can. To ensure a precise result, retracing their letters or symbols, they had to replace the existing steering mechanism by a more accurate servo.

The biggest problem at this day, was caused by the energy management. The housing and the self constructed mechanism were very heavy and a lot of energy was needed to st the object in motion.


Group 4

Working on their flowering pot, they started their final model today. For the outer housing, they use a bucket from the construction market. Taking off the bottom if, the chassis of the remote controlled car should sit inside of the bucket and be covered by it. The inner part is a ready made flower pot. Four light sensors are mounted around it, to get a very precise result. The sensor for the humidity of the flowering soil is mounted already.

Also the programming is on a good way. Having tested the threshold for light and humidity, the first methods and steering instruction are finished and could be tested with the engines. The big challenge they see, is the weakness of the engines for the big final model.


Group 5

IMG_8622 IMG_8624 Video

Group 5 gave up the idea to use a toy dog as „housing“ of their mechanical construction the day before. The toy was delivered at thursday and was much to small for their construction. So they started to build a two dimensional silhoutte of the dog, which should be mounted to both sides of the construction. To check the possible, resulting features, they worked first with a cut-out paper model and simulated the positions of the hinges (what can be seen in the video). After this low-resolution prototype they used the laser-cutter for the final model. At this time, they want their dog to „sit“ and to „lay down“ – and, if there is some time left, to „sit up“.

Also the software is already on a high level – using voice control to command their dog. The biggest challenge in this moment were the mechanics. Working with two servos, there are high expectations to the precision of the construction and model building.


Group 6


Having spent much time on the design of the housing, the upper part of the monster is already finished and also the rest of the housing is very detailed and on a high level. Troubles were caused by the mechanical parts, like the pressure sensor,  which is not always able to detect if the monster was fed with Maoam. With the housing on this level and the software, which is also already half-finished, the team is very optimistic to hit the deadline.

linked categories 2014b, Daily Logs


Write a comment