Introduction

• We have a rectangular maze divided into cells, where each cell is either a wall or an open space. The maze is closed, meaning that it is not possible to pass through its outer bounds.
• The "capture the flag" game is played with one or more robots. Each robot starts on a different cell; this initial cell is the robot's home-base.
• A robot can move to the cells next to him (north, south, east, west) if there is no blocking object, like another robot or a wall.
• Somewhere in the maze on one of the cells, is a flag that the robots are trying to capture. To capture the flag, a robot must be standing on the same cell as the flag. Once a robot captures the flag, all the other robots are informed. The robot that brings the flag back to his home base is the winner of the game.
• The only chance to take the flag away from a robot is the earthquake button. Like the flag, the button is located on a random cell somewhere in the maze and can be pressed by a robot standing on that cell. When the earthquake button is pressed, the flag and the earthquake button positions are changed to new random positions and all robots return to their home-base. The earthquake button only works after the flag has been captured.
• The robots have a special feature that allows them to pass an inner wall in the maze: They can build a bridge. Building a bridge allows the robots to transform a wall cell into a cell that the robot can walk through, but it can only be used by the robot that built it. Since building a bridge is a hard task, the world starts off with an initial number of pre-built bridges and when this limit is used, a robot is delayed for a certain amount of time while the requested bridge it built.
• In order to navigate, a robot is able to sense the cells next to him in each of the four directions (north, south, east, west) and for the cell he is standing on (center) the following information:
• The type of the cell  (if it is a wall, a robot, a bridge of another robot, his own bridge, the earthquake button, or the flag).
• The direct distance to the flag (not necessarily an actual path to the flag).
• The direct distance to the earthquake button (not necessarily an actual path to the button).

Implementation

• WorldService: This interface defines the game administration methods; it is implemented by the WorldServiceActivator class. The WorldServiceActivator class not only provides WorldService methods, but it acts as the core of the World interface implementation. WorldServiceActivator contains all of the game state information and the methods for the actions that the robots can perform. There is only one instance of the WorldServiceActivator class in the game, but every robot has its own instance of the World implementation. The robots do not use the shared WorldServiceActivator object directly, instead they call methods on their own World object; in this case, the World implementation is merely a wrapper where all World methods delegate their calls to the WorldServiceActivator instance. This approach keeps the interfaces clean since robots do not need to pass in an identifier or anything to WorldServiceActivator and they also cannot access the the adminstrative methods in the WorldService interface.

• Robot: This interface defines the methods used to manipulate a robot. The robot implementation also contains the algorithm that makes the robot perform its actions on its World. The main goal of the robot algorithm is to find the flag and return it to the robot's home, but it may also define some strategy for dealing with the earthquake button and building bridges. A robot implementation is notified when another robot captures the flag; this occurs when the Robot.flagCapturedHandler() callback method is invoked; this gives your robot the possibility to react on such an event. You may want will most likely need to keep track of internal data about the maze to help it navigate and make decisions.