Archive for the ‘bots’ Category
Are swarms chaotic and why should you care?
The zoologist and his colleagues discovered that when a swarm contains between 25 and 74 locusts per square metre, the locusts are almost always aligned but exhibit rapid and spontaneous changes in direction. There were almost no directional changes above that range of densities. [read more Sychronising the swarm]
So what does this mean for software and robotic swarms? Are there densities of particles in the swarm that will drastically change the behavior? Are swarm systems chaotic and if so what does that mean for designers of swarms? Many of the same simple rules we are using to program our swarms are based on insect swarms. So it isn’t a far reach to think that perhaps some of the odder previously thought to be unexplained behavior of insect swarms may show up in our software and our robot swarms at specific densities.
Swarms are being used in more real world applications. What happens if the swarm balancing network traffic on your server all aligns and sends all the traffic to one machine? Or more troubling what happens when US Military swarm based robots suddenly evolve new unexpected behaviors?
This is clearly an area that needs serious research in the very near future.
All is not bad, we can also use studies like this to predict crowd behavior in crowded situations using swarm models. Knowing at what crowd density the behavior changes can help us better design buildings and infrastructure preventing tragedies like the Rhode Island night club fire a few years back.
More information:
Boids ( Flocks, Herds and Schools: a Distributed Behavior Model )
The Application of Computational Models for the Simulation of Large-Scale Evacuations following Infrastructure Failures and Terrorists Events
Swarm-bots project
Why Locusts Abandon a Solitary Life for the Swarm
CyberLover swindles the willing
Well this was just a matter of time. Many chat bots have been trolling the internet doing a very good job of passing themselves off as humans. The easiest way to do this is of course to find a willing victim. CyberLover does just that.
CyberLover is fast and loose establishing up to 10 relationships every half hour. CyberLover then extracts personal information from the unwitting but willing victim.
Some of the best computer hacks done have been through social engineering rather than with clever code. This brings the best of both worlds together.
Right now CyberLover has been targeting Russians almost exclusively, the rest of the world is expected to be targeted in time for Valentine’s Day.
Personal agents may reach your phone before they reach your computer
Several companies are working on intelligent agents to help you in your computer interactions, ( see Lua created assistant ) and here is yet another one, Magitti.
Magitti is different in that it is taking advantage of some of the phones unique advantages: portability, gps and that it travels with you all day.
It is also different in that there is a built in revenue stream. Magitti recommends local restaurants and shopping at appropriate times and locations.
There are few details to be found at PARC yet. More will be forthcoming I’m sure. There are papers available for a fee at IEEE search for ‘PARC and mobile’.
See also:
