Online least-squares policy iteration for reinforcement learning control | IEEE Conference Publication | IEEE Xplore