PLATO: Policy learning using adaptive trajectory optimization | IEEE Conference Publication | IEEE Xplore