Robotarium X - It's Full Of Robots!!!

Intelligent Machines has a nice post upon a new kind of zoo: A robot zoo! See the complete story here.

You can watch a video here.

The Genetic Arm 2.0

Martin Pelikan, from MEDAL Blogging, has a nice post about a neural network, which is trained through a genetic algorithm, to throw a ball as far as possible. See here. There is even a video of the arm throwing a ball.

The Watchmaker Framework for Evolutionary Computation

There is an interesting Evolutionary Computation project which uses Java to implement a general platform-independent software to run Evolutionary Algorithms. See here.

The authors have two nice demos of their software:

1. The Traveling Salesman Problem.

2. Interactive Evolution of Biomorphs.

You can also find some tutorials upon optimization and Evolutionary Computation:

1. Tutorial Part 1 - A Brief Introduction to Evolutionary Algorithms.

2. Tutorial Part 2 - Implementing Evolutionary Algorithms in Java with the Watchmaker Framework.

3. Performance Tuning & Optimisation (incomplete).

Amir's View On Scholarpedia

There is a nice post from Thesilog upon Scholarpedia. See here.

It seems that our blog friend, Amir, is coming back to the blogging life again! :)

Take a minute and read his interesting view about the subject!

Swarm Of Smart Rockets

Martin Pelikan, from MEDAL Blogging, has a nice post upon the evolution of small rockets through evolutionary computation, more specifically, genetic algorithms. See here.

The genetic algorithm seeks to optimize fuel consumption, navigation, and engine features.

Evolving Lego Structures Through Simulated Evolution

There is an interesting post at Think Artificial upon building Lego structures through evolutionary algorithms. See here.

There are nice structures evolved: bridges, trees, cranes, etc.

The simulator even takes into account some physical parameters, such as joint's configurations and gravitational forces to decide which blocks need to be prunned off or added. It's a nice post!

Ingo Rechenberg and his team at Technical University of Berlin has also applied evolutionary algorithms (a Nested Evolution Strategy) to evolve structures, such as bridges (more bridges here), cantlevers, and lens. For more applications, see Evolution Strategy in Action: 10 ES-Demonstrations. Hans-Paul Schwefel applied an evolution strategy to evolve a two phase flashing nozzle.

See below some images of structures evolved by Rechenberg's team.

01. A bridge.

Initial Configuration:



Final Evolved Configuration:




02. Lens.

Initial Configuration:




Final Evolved Configuration: