Wenguo and I arrived yesterday with a swarm of e-puck robots. Wenguo and Jan Dyre have done a sterling job programming the e-pucks with 3 simple behavioural rules:
if you see an obstacle turn to avoid it, else
if you see the tail lights of another robot, follow it, else
go straight ahead.
The robots will thus form an orderly queue, except that because the arena is too small the line very quickly gets broken when the lead robot encounters a wall. However, that all just makes for an interesting – and constantly changing – swarm.
“COOL”: definitely the response Vivek Murali Thiyagarasan wanted when going out on to the floor at @Bristol. There was an interactive presentation on robots; what are they? what are they used for? where are robots? All sorts of questions were flying around, in all directions – “what’s that?”; “wow, is that?”
Several different demonstrations which not only made participants of all ages think again about robots, and Artificial Intelligence but posed further questions which they could take away. Many @Bristol staff were around to help the researchers and even they were intrigued. “A very impressive collection of acts” says John; “lots of interesting experiments”.
Oliver Bown had a computer making music. Each kid joyfully says “robots can make music” after jumping straight in-front of the exhibit. Not only were the kids amazed at all manners of music coming out of the computer, but so was Oliver …
Henry Marten and David McGoran decided to walk round as a robot today. Challenging people to ask ‘What is a robot?’ and ‘Why is that guy walking around in a cardboard suit?’ Answers to the first question included: ‘Because his legs are showing’ and ‘He has muscles.’ Answers to the second question were mainly ‘Because I can tap him on the shoulder and run away!’ All in all our cardboard robot proved so successful extra researchers were drafted in to answer a barrage of questions.
For the second session Henry is planning a few modifications to make it a little more comfortable and, I reckon, giving him a chance to turn round faster.
One of our number, Nathaniel Poate, is a minor celebrity. His long road into the field of robotics started with an appearance on robot wars. He can be seen here in this youtube video discussing the advantages (a powerful spike) and disadvantages (weak armour) of his robot Short Circuit. After a successful start he was tragically sent to the pit by Hypno Disk. Fortunately, that wasn’t enough to put him off the whole idea and he continues to build robots to this day.
What’s next? The time is consuming, the beginning of the end is coming and we can find @Bristol walking robots…
Bigger or smaller, it doesn’t matter; these wooden machines can walk around @Bristol among real people… These ones work with a couple motors, sensors and some wires which represents the muscles, senses and nerves respectively.
The 60-year old Dutch artist Theo Jansen builds gigantic structures that can walk over the beach of The Netherlands. The way he does this is very unique, in that he uses a genetic algorithm to design the best set-up. What this means is that he does not type any code himself, but let the robots walk over the beach and let them combine the best options to find one that is even better, similar as in evolution.
Let’s see how this looks:
All of these creations are made of 375 plastic tubes, costing only 10cents per meter, making it very light and inexpensive. It is held together by cables, nylon strings and adhesive tape. By trial and error the ideal settings are found to walk as stable as possible over the windy beach. All of the creations are driven by energy from the wind.
The first creations were simulated in a computer program, before building the first real one. These required a lot of tuning (through genetic learning) before they were able to stand. Next versions had to learn how to deal with the strong wind.
This way he has designed a new nature, that is able to evolve using survival of the fittest. Eventually he wants a large number of these creatures walking around on the beach, interacting with eachother, without his interference.
The subtle manipulation of objects is a pressing challenge in robotics. To convey the problems researchers face Martin Postler is using this workshop to create a glove to restrict movement. This forces people to think about the complexities present in apparently simple tasks (pictured below) and highlights the importance of good design within robotics.
It also raises the interesting question: To what extent is intelligence embodied? If we could build a robot with the physical senses and capabilities of a human being (artificial muscles, an incredibly sensitive sense of touch etc…), how much intelligence would we need? Personally, I feel that quite a lot or AI within robotics is developed to overcome the robots own physical limitations. As increasingly sophisticated engineering solutions become available we may discover that intelligence is simpler than we realised.
Wenguo and I arrived yesterday with a swarm of e-puck robots. Wenguo and Jan Dyre have done a sterling job programming the e-pucks with 3 simple behavioural rules:
Recent Comments