Quantum Algorithm to Solve a Maze: Converting the Maze Problem into a Search Problem

Abstract: We propose a different methodology towards approaching a Maze problem. We convert the problem into a Quantum Search Problem (QSP), and its solutions are sought for using the iterative Grover's Search Algorithm. Though the category of mazes we are looking at are of the NP complete class, we have redirected such a NP complete problem into a QSP. Our solution deals with two dimensional perfect mazes with no closed loops. We encode all possible individual paths from the starting point of the maze into a quantum register. A quantum fitness operator applied on the register encodes each individual with its fitness value. We propose an oracle design which marks all the individuals above a certain fitness value and use the Grover search algorithm to find one of the marked states. Iterating over this method, we approach towards the optimum solution.