This research aims to develop an UAV path planning in air corridor using deep RL. The problems with UAV’s would be huge energy consumption, limited battery life, delivery time and drone collisions with obstacles or with each other. By utilizing the potential of Convolutional Neural Network(CNN) especially Reinforcement Learning and Generative Adversarial Network, we plan to enhance the capability of UAVs to deliver freight in a faster, energy efficient and safer way. This research also assesses if near real-time performance can be achieved for such an approach using EEG signals. This research idea is useful in many domains like cargo delivery, unmanned driving, medicine delivery, e-commerce etc.