Basic simulation and evaluation

Random search

This is the brain of our pods. It makes our pods "think" and "come up" with trajectory plans for the near future. To make it work, we need to try a lot of different trajectories and evaluate the outcomes they provide to pick the trajectory with the best score. Trajectory is defined with a list of game moves for the next N turns. We can easily generate trajectories randomly and then simulate the game with them:

Solution sol = new Solution(); float score = sol->RandomizeAndEvaluate(pods);

then just run this line 10 000 times more and if score of a solution is more than the best found then save it