Output-feedback Quadratic Tracking Control of Continuous-time Systems by Using Off-policy Reinforcement Learning with Neural Networks Observer | IEEE Conference Publication | IEEE Xplore