Bayesian reinforcement learning in continuous POMDPs with application to robot navigation | IEEE Conference Publication | IEEE Xplore