Use the lidar to calculate adversaries trajectories

[Terae]

Actually, the Lidar code is able to return a vector<repere::Position> object which represents all others robots.

It works without memory and calculates at each frame the time-and-space average of points to approximate adversary robots.

A relative easy-update would be to memorize the N previous frames and make some new processes:

• Determine the relation-ship of robots between 2 consecutive frames: determine which robot of the first one correspond to which of the second one; or else if a robot (dis)appears --> tag each robot with its position among frames
• Provide some helpful getters from these data, for example accessing the average speed vector from the last N seconds/frames and so we should be able to anticipate adversary moves

This update could improve our strategy by detecting in advance where adversaries want to go, and we can also distinguish our secondary robot from all detected robots by tracking it, and so approximately know its task.