Autonomous Obstacle Avoidance Robot Using Reinforcement
Abstract
This research work focus to build an autonomous obstacle avoidance vehicle using reinforcement learning. The research work proposes a Reinforcement Learning based approach to train the vehicle agent, such that it can intelligently traverse any given terrain under diverse and even unforeseen scenarios. This vehicle is primarily built using a Raspberry Pi and ultrasonic sensors. The ultrasonic sensors detect distance from the obstacle and communicates the data to the microprocessor. This data is then used to predict the next state of the vehicle using Reinforcement Learning. Further, there is scope to extend the project towards a GPS enabled autonomous system for localization and navigation