* PYTHONPATH = group1/

* Agent: GerdLorenzoAgentFactory

* Agents.py : Concerning BasicScoringAgentFactory, you can make your personal AgentFactory by deriving from BasicScoringAgentFactory and overwriting its constructor. New Factory constructs two ScoringAgents with scoring function as specified in the custom constructor (by Sebastian + Jan)

* test_Agents.py: testing of standard beginning agent (by Gerd + Lorenzo) (currently tests tested ScoringAgents.__init__, ScoringAgents.choose_action)

* get_distance([x1,y1],[x2,y2]): returning the distance of any two points of the current maze (by Andrei + Peter)

* single_nearest_food_feature(self,position,game_state,agentIndex): returning 1./distance of nearest food. Alternative feature: distance/longest_Manhattan_distance

* single_nearest_friend_feature(self,position,game_state,agentIndex): returning 1.-1./distance of nearest friend - currently only using MH-distance, so it is seriously flawed!!

* single_nearest_enemy_feature(self,position,game_state,agentIndex): returning 1.-1./distance or 1./distance of nearest enemy depending on whether our agent is in his half or not - currently only using MH-distance, so it is seriously flawed!!

* ManhattanLunchtimeAgent: implement agent based on manhattan distance going only for food (by Sebastian + Jan)

* ScoringAgent: implement agent based on scoring of an abirtrary number of feature functions.

* test_ManhattanLunchtimeAgent: prepare a larger test-mazes (with / wo walls) to test the ManhattanLunchtimeAgent (by Gerd + Lorenzo)

* function+interface still to be defined: initialize maze during start-up, prepare look-up-table for distances in maze. (by Andrei + Peter)

* dc.get_nearest_friend_distance(game_state,agent_index,position)

* dc.get_nearest_enemy_distance(game_state,agent_index,position)

* dc.get_all_food_distances(game_state,agent_index,position)

* dc.get_all_friend_distances(game_state,agent_index,position)

* dc.get_all_enemy_distances(game_state,agent_index,position)