Towards modeling the learning process of aviators using deep reinforcement learning
Joost van Oijen, Gerald Poppinga, Olaf Brouwer, Andi Aliko, Jan Joris Roessingh · 2017
In this paper we report on our study of the performance of Deep Reinforcement Learning (DRL) agents in performing tasks that are illustrative for human Sensor Operators (SOs) in Remotely Piloted Aircraft Systems (RPASs). Our hypothesis is that the descriptive and predictive qualities of the agent's learning process potentially allow us to identify human task requirements, training needs, selection criteria and cut-off benchmarks. We present DRL results on tasks that cover different cognitive abilities required for an SO, using games as a method for learning.