The coursework for CS118, Programming for Computer Scientists, was to create a bot that solves a maze, in java. The final coursework must have a bot that solves the maze, and remembers the route it took to complete the maze, such that if the bot was run on the same maze twice in a row, it would reach the end without backtracking. This did not involve storing the route long-term in anyway.
I will not go into great detail with my solution for this, as the coursework is reused each year, unlike CS141, and in the unlikely event some clueless fresher is reading, I don't want to spoil their fun. I will, however, say that my solution used an optimised DFS, rather than the BFS or Asolutions that others gave. My optimisation gave me results close to an optimal solution, with minimal backtracking on the first run, meaning it's first time trying to solve a maze was considerably more efficient than those using solutions like A, though my route is not garunteed to be the most efficient.