Safe exploration of state and action spaces in reinforcement learning
Journal of Artificial Intelligence Research
Hi-index | 0.00 |
In this paper we present a method to obtain a near optimal neuro-controller for the autonomous helicopter flight by means of an ad hoc evolutionary reinforcement learning method. The method presented here was developed for the Second Annual Reinforcement Learning Competition (RL2008) held in Helsinki-Finland. The present work uses a Helicopter Hovering simulator created in the Stanford University that simulates a Radio Control XCell Tempest helicopter in the flight regime close to hover. The objective of the controller is to hover the helicopter by manipulating four continuous control actions based on a 12-dimensional state space.