[Tweets]

13/07/2018 11:42am
Artificial Consciousness: Is it possible to give a #robot a soul? [LINK]
12/07/2018 6:36pm
RT @DeepMindAI: Measuring abstract reasoning in neural networks - our latest #ICML2018 paper - takes inspiration from human IQ tests to exp…
10/07/2018 2:15pm
Preferential #foraging leads to a better #swarm performance in dynamic environments, provided that information flow… [LINK]
06/07/2018 10:19am
Personalized “deep learning” equips robots for autism therapy @MIT via @Robohub It's so great that through technolo… [LINK]
05/07/2018 8:27pm
Looking forward to presenting our paper on information flow regulation in preferentially #foraging #robot #swarms a… [LINK]

[Linee the line follower]

Robots Date: Dec 2013

Linee the line following robot Linee is a line following robot that instead of being programmed, uses a fully analogue controller board. It features headlights, light sensors, a comparator chip, wheels and other small parts that make it follow either black lines on a light background or white lines on a dark background. This is my first real robotic project ever.

How does it work? The resistance of the light sensors (photo resistors) mounted in the front, facing the floor, is different on the two sides when the robot is crossing a line (i.e. when a different light reflection due to a different colour is being perceived). At that point, the comparator chip finds the difference and sends a higher voltage to the proper wheel, making it turn faster, so that the robot turns slightly until the colour perceived by the sensors is the same again. The green or yellow indicator LEDs light up as well to signalise which side is getting power. This way of processing information makes the robot ‘wiggle’ as it follows a line.

The robot consists of two soldered breadboards. The sensory breadboard is in the front and contains variable resistors that can adjust the strength of the headlights and the balance between the right and left side of the light sensors. The control board sits on top of the robot and connects the sensory board with all the other necessary electronics, wheels and the power supply. The boards are mounted on a basic mobile chassis with custom-built body made of Meccano parts. There is also a switch that changes the behaviour between black-line-following and white-line-following.



Control board top Sensory board top
Control board top Sensory board top
Control board bottom Sensory board bottom Empty chassis
Control board bottom Sensory board bottom Empty chassis


{Please enable JavaScript in order to post comments}

Ilidian the robotic dinosaur

Meet Ilidian, my robotic dinosaur pet. He is a male Pleo rb, in my opinion one of the most advanced home robotic pets at the moment. He has a lot of touch sensors so he can feel being stroked or hit on various parts of his body, temperature sensors that allow him to feel dizzy when hot or shiver when cold and much more stuff that make him quite life-like.

LED Christmas Tree

Build Brighton, a group of people interested in electronics, robots and a lot of other stuff, have created electronic Christmas kits and are running workshops where you can learn how to solder the components together. The kits include a Christmas tree and a snowman badge, both with blinking LEDs.

Il Matto: Hello World

Il Matto is a microcontroller development board that uses Atmel's ATMega644PA AVR chip. It was developed at the University of Southampton, which means that I get to play with it thanks to Klaus-Peter Zauner.

NXTPong

A custom-built lego machine for playing pong, featuring two controllers connected to the Mindstorms NXT brick. Programmed using NXC.

pyCreeper

The main purpose of pyCreeper is to wrap tens of lines of python code, required to produce graphs that look good for a publication, into functions. It takes away your need to understand various quirks of matplotlib and gives you back ready-to-use and well-documented code.

Information Flow Regulation in Preferentially Foraging Swarms

Swarms are tasked with foraging from multiple sources in dynamic environments where source locations change periodically. Preferential foraging leads to a better swarm performance, provided that information flow among robots, that results from recruitment, is regulated