A coursework problem: three video cameras, mounted on the ceiling of a room, need to be positioned to maximize the total floor area they cover between them. Built in Python, comparing three search algorithms against each other on the same problem instance.

The three algorithms

Hill Climbing without retrace, Hill Climbing with retrace, and Simulated Annealing — all applied to the same coverage-maximization problem. Simulated Annealing came out ahead: it found the maximum coverage area and camera positions in a few minutes on the example instance, where the Hill Climbing variants were more prone to getting stuck.

Took about two weeks end to end, and scored a 10/10 on the exam.