Raspberry Pi - Introduktion af indlæg


Raspberry Pi v. Alan Mycroft

During the past 10-20 years computer systems have become both ubiquitous and more powerful, but also more closed. By 2008 children increasingly had mobile phones with more computing power than the PCs we learned to program on, but could not program themselves – university computer science applications reduced in quality and quantity. Example barriers include fashionable laptops lacking screws to look at the internals, or a child installing Linux wrecking a family’s PC.

The Raspberry Pi was a reaction – instead of a gadget whose internal workings remain a mystery, we wanted a fully fledged computer at the price of a toy (or a textbook).

The initial Raspberry Pi exploited the “annually more transistors for the same price” Moore’s law to build an ARM, HDMI graphics card and 256MB of memory on a $25 credit-card-sized board with GPIO pins exposed, with Linux as operating system – exposing python, scratch and bare-metal programming.

But Moore's law continues with the current Pi 3 having 1GB of memory, a quad-core processor and Wifi. The results have been remarkable – nearly one million were sold in the first year and sales recently reached ten million – but everywhere people are doing newly interesting things with the hardware (e.g. putting one on a weather balloon, controlling robots, adding cameras/touch-screens, even networking them as cheap supercomputers).

“Raspberry Jams” have sprung up when children once again play with, and program, computers for fun. School curricula have incorporated Computing as a life skill (Jeanette Wing's “Computational Thinking”); the effect is better students for us and thereby better innovators for society.

Raspberry Pi? Men vi har jo iPads! v. Karin Høgh, PodConsult

Midt i en tid, hvor skoler og familier anskaffer iPads og bærbare computere, dukker Raspberry Pi op. Hvorfor nu det? Hvad kan den bruges til i undervisningen? Er den ikke for besværlig? Er det ikke kun for nørder?

Karin Høgh forhandler læringsteknologier til familier, skoler og videregående uddannelser. Hun producerer og sælger RaspberryPod, verdens første komplette sæt omkring Raspberry Pi. Målet med RaspberryPod er at gøre det nemt og håndterbart for lærere og skoler at bygge en computer fra bunden, lære om tilslutninger og elektroniske dele, kode i Python og derved gøre det muligt for de elever og studerende, som har interessen, at tage den derfra og videre ud i spændende jobs som programmører.

Python på Raspberry Pi v. lektor Thomas Arildsen

Python er et særdeles populært programmeringssprog, som i disse år bl.a. bliver mere og mere populært til videnskabelige beregninger, men som henvender sig bredt til mange forskellige anvendelser og brugergrupper. Python er bl.a. populært pga. sin simple opbygning, der gør det let at lære, og sit kæmpe udvalg af udvidelsespakker, som blomstrer ud det store og aktive udviklermiljø, der findes omkring
Python. Her gives et kort overblik over Python og nogle af de muligheder, der er i kombinationen med Raspberry Pi. Vi ser bl.a. på softwarepakker, der giver nem adgang til I/O på Raspberry Pi.

Konstruktion omkring en Raspberry Pi v. lektor Søren Krarup Olesen

Gennem en årrække har folk efterspurgt at kunne måle lyd med samme nemhed, som man eksempelvis aflæser sit termometer på væggen. En app på en smartphone lyder derfor som det oplagte valg. Vi vil i indlægget se på, hvorfor Aalborg Universitet alligevel fravalgte app-løsningen og i stedet valgte at konstruere et decideret stand-alone instrument, som er udviklet omkring en Raspberry Pi. I udviklingen blev der især fokuseret på kalibrering, reproducerbarhed og intuitiv brug.

The Casting Demonstrator v. lektor Petur Olsen

The demonstrator is used to showcase the floor heating control algorithms that have been developed the European research project, CASSTING. The demonstrator uses a large screen to show the floor plan of the building. The dynamics of the floor heating system are modelled in Matlab/Simulink. To enable both simulated and physical devices the Homeport system is used. Homeport enables a homogeneous interface to heterogeneous devices. This way the interfaces to simulated and physical devices and be abstracted and accessed in a uniform way.

The graphics, Homeport, and the simulation all run on a Raspberry Pi. Graphics are developed in SDL and C++, Homeport is developed in C, while simulations are implemented with code generation through Matlab Coder. The experiments show that the Matlab/Simulink code and Homeport can easily run on the Raspberry Pi, using about 40% CPU. Unfortunately, the graphics power is lacking. The Raspberry Pi can only managing about 2-4 frames per seconds on the Full HD screen. This is enough for the slowly developing floor heating system, but for more reactive systems it might not be possible. Newer versions of the Raspberry Pi offer new and more powerful CPU, but offer the same graphics chip, so upgrading to a newer version would not help solve our issues.

Kontakt

Facilitator
Jørgen Biegel
TEKKVA CONSULT
Tlf.: +45 96 35 45 65
E-mail

     
InfinIT er finansieret af en bevilling fra Styrelsen for Forskning og Innovation og drives af et konsortium bestående af:
Alexandra Instituttet . BrainsBusiness . CISS . Datalogisk Institut, Københavns Universitet . DELTA . DTU Compute, Danmarks Tekniske Universitet . Institut for Datalogi, Aarhus Universitet . IT-Universitetet . Knowledge Lab, Syddansk Universitet . Væksthus Hovedstadsregionen . Aalborg Universitet