#1 Human-like Bot: A competition for CEC2011
The CEC2011 "Human-like Bot" competition is similar to the BotPrize, so competitors can use it as a warm-up for the 2011 BotPrize (not announced yet, but expected to be part of CIG'11 in Soeul, Korea, in September 2011).
Your challenge is to program a bot to play the FPS Unreal Tournament 2004, and to make it behave like a human player. Judges will play games against your bot and try to tell if it's human or bot. The bot that does best at fooling the judges will be the winner.
Full Details can be found at: http://www.botprize.org/HumanlikeBots.html
#2 Testing Evolutionary Algorithms on Real-world Numerical Optimization Problems
Over the past few decades several efficient Evolutionary Computing (EC) algorithms have been devised for handling large scale optimization problems. Performance has mainly been demonstrated using benchmark function sets with known optima. However, these performance benchmarks may not guarantee a similar performance on practical optimization problems.
This competition aims at bridging the gap between EC algorithms and real-world problems, characterized by objective functions derived from a practical problem. Researchers from academia and industries are invited to submit unpublished manuscripts applying state-of-the-art nature-inspired optimizers to solve documented tough problems.
Full details can be found at: http://www3.ntu.edu.sg/home/epnsugan/index_files/CEC11-RWP/CEC11-RWP.htm
Ms Pac-Man, one of the most successful arcade games ever, is a challenging real-time video game where Ms Pac-Man, controlled by the player using a four-way joystick, has to collect pills for points and, at the same time, avoid to be eaten by any of the four ghosts chasing her.
Traditionally, the game is viewed from the perspective of Ms Pac-Man and several competitions have taken place in the past where controllers were developed to direct Ms Pac-Man throughout the game. However, the game also provides an excellent environment for testing multi-agent strategies by controlling the team of ghosts; here the goal is to minimise the overall score achieved by Ms Pac-Man.
An efficient simulator of the game has been developed for the purpose of this competition, allowing contestants to connect their controllers using simple software interfaces. This year, for the first time, contestants may provide controllers for either Ms Pac-Man and the Ghost Team or both; in the latter case, collusions between the controllers is prohibited.
The full details of the competition, including source code and tutorials, may be found at: http://www.pacman-vs-ghosts.net
The travelling salesperson problem is one of the most widely studied optimisation problems, and there are a plethora of clever algorithms and meta-heuristics for finding good solutions. Close to optimal routes can now be found for problems involving thousands of cities. The objective in the standard TSP is to minimise the total distance travelled. The PTSP adds a simple twist that has far-reaching effects: the salesman has mass, and moves by choosing a force vector to apply to the mass at each point in time.
The objective now is to minimise the number of time steps taken to visit all the cities. If two solutions tie on this, the tie-breaker is the solution that uses the least force.
Full details can be found at: http://algoval.essex.ac.uk/ptsp/ptsp.html