Blog - Sketching with Hardware

Fridolin Fikus

Published on: | Author: Sketching With Hardware | Categories: 2014b, Best Projects, Projects

Our Team

Team 4 consisted of Max Hackenschmied and Nicki Schäfer. Max is doing his Master Degrees in Human-Computer-Interaction and has already a lot of experience in programming, Nicki is studying Industrial Design at the Technical University in Munich and a complete beginner in programming or hacking stuff.

 

IMG_8590

Our Topic

After we came up with the entertaining wing-controller for Flappy Bird at the first day, we also wanted to spend the rest of the project to develop and design an amusing and perhaps entertaining object. The brainstorming session at tuesday brought us to the idea of Fridolin, the intelligent and ‘living’ Fikus.

Everybody knows the problem of house plants, which are forgotten after the first weeks and drying out in the corners of our rooms. We wanted to use the chassis of our remote controlled car to give something like an „own personality“ to our home plant Fridolin. Measuring the conditions of Fridolins environment with different sensors, the Arduino controlled chassis should carry Fridolin through the appartment and look for a spot with perfect growing conditions. So Fridolin should start to look for a sunny place „on its own“, after standing in the shadow for some time. The flower pot also should check the humidity and start to attract attention, if Fridolin has been forgotten to be watered. In summary, we wanted to give something like a human existence and human behaviours to our home plants, which are not able to to move or talk… up to now.

 

Key Features

IMG_8604-2

We started to work out our concepts with some rough prototypes and mockups. To measure the conditions of the environment, we wanted to work with light sensors, to see if Fridolin has to „get up“ and look for a place with better illumination. With a humidity sensor, we wanted to check if the plant has to be watered. And with distance sensors in the front and in the back of the chassis, we wanted to avoid collisions with walls or funiture, while Fridolin is taking his trips through the appartment. In addition to this, we wanted to use LEDs to show to the user if Fridolin is awake and active or not.

 

Working with Sensors

The work with the light sensors went pretty well. We installed four of them around the inner flower pot, to compare the values and to be sure, that Fridolins own shadow does not cause unnecessary trips. We later defined a threshold value in the source code, falling below this value, caused Fridolin to find a new place.

Therefore, the humidity sensor was a bigger challenge. Because, there is no standard Arduino sensor to measure the humidity, we had to build and define one on our own. We checked some sources in the internet for necessary informations and started our own series of test. Out of two nails, mounted with a defined distance to each other, we built our own sensor and tested it with three different levels of watered potting soil. The soil between the nails works as an electrical resistance, which gets bigger as soon as the potting soil dries out. So, with the three different levels of watered potting soil, we were able to define a field, in which the humidity of Fridolins potting soil should be best.

IMG_8610 IMG_8608

Also the mounting of the infrared distance sensors was a big challenge. We wanted to mount them at the bottom of our construction, to hide them as much as possible. But in this position the sensors detected interfering signals, which were caused already by small changes of the floor. One possibility to solve this problem would have been, to work with an average value, to ignore temporary peaks. Unfortunately, this was not possible, because Arduino is not able to work with multiple threads at the same time – so the time between the detection of a barrier and the collision would have been to small. We could solve the problem mainly manual, by adjusting the sensors in slight angle to the top – so the interruption signals, which were mostly caused by the floor, were not detected anymore.

 

Software Development

IMG_8641

As Arduino doesn’t support multi-threading, pseudo-parallel tasks are quite challenging. Fridolin has to check several sensors, manage driving, play sounds and react to user-controls via Bluetooth at almost the same time. Moreover, their are several tasks based on time. In order to deal with that, we employed our own Tick()-function.

Also the driving logic was a bit complicated. We soon realized, that the two infrared-sensors weren’t ideal for collision detection, but the lack of alternatives forced us to use them. As he is quite wide and the sensors’ viewing-angles are narrow, Fridolin sometimes behaves like a child not knowing how fat it realy is. So edges, small gaps and so on are a bit problematic.

 

Model Building

IMG_8618 IMG_8617

For the final model we worked with customized ready made parts. For the outer housing, we used a bucket out of the construction market. The inner part, was a „real“ flower pot made of plastic. To fill the gap between the two parts, we cutted a ring of black plastics and used the surface to mount the light sensors in it. To hide our chassis, we took off the bottom of the bucket and built a new one, inside of the housing. So, when the housing was put on the chassis, it „floated“ some millimeters over the floor.

 

Finish

IMG_8639 IMG_8642

In the last steps, we placed the LEDs also in the lower part of the bucket, to show with the reflected light on the floor, if Fridolin is awake. After wiring the components and making the last fine adjustments in the night before the end presentation, we ware able to plant Fridolin in its new pot, at Tuesday before the presentation. Though we had tried to keep the weight as low as possible, the energy management became a big problem in the last hours before the presentation. With the weight of the plant, the engines came to their limits. We could solve this problem mainly, by empowering the engines with three more batteries, but, unfortunately, the resulted driving experience was not as smooth as we had imagined.

 

IMG_8648

 

Video:Video – Fridolin Fikus

In summary, we had a very good time and we like the result of Fridolin Fikus very much. Even we had some mechanical problems in the last day and hours, the workshop was very interesting and motivating. It was quite fun, to work on this topic and to try out things and technics in such a new contexts – and of course, the support and the motivation of the tutors was great. Thanks for that!

linked categories 2014b, Best Projects, Projects

|

Write a comment

Comment