Maze Solving Using Deep Q-Network
Anushtup Nandy, Subash Seshathri, Abhishek Sarkar · 2023
Path planning and obstacle avoidance are crucial for enabling the autonomy of mobile robots to operate in real-world environments. Conventional algorithms are known to be computationally expensive, and they require prior knowledge of the environment. In this paper, instead of using conventional algorithms, we present the usage of DQN (Deep Q-Network), a reinforcement learning algorithm, to solve the path planning problem. The goal was to observe and report the feasibility of using DQN as a path-planning algorithm for mobile robots in maze environments with walls leading to dead-ends. We have also showed the possibility of improvements to the algorithm, which can be used in more challenging maze environments.