Pragmatic-Implementation-of-Reinforcement-Algorithms-For-Path-Finding-On-Raspberry-Pi

In this project, we did a pragmatic implementation of an indoor autonomous delivery system that exploits Q-learning and Deep Q-learning which are reinforcement learning algorithms,for path-finding and obstacle avoidance in a given environment. The main computation is done on the device and the best possible path is hosted online using an API which is then extracted by the Raspberry Pi module. The implementation is done using motor driver and encoders. The proposed system is a moderately cost-efficient approach that is implemented to compute the shortest path between the source key point and destination key point to carry out the desired delivery and then navigate through the path by avoiding the obstacles in the path.