Discussion and Conclusion

Three different control schemes were studied in the framework of nonlinear state-space models. Direct control is fast to use, but requires the learning of a policy mapping, which is hard to do well. Optimistic inference control is a novel method based on Bayesian inference answering the question: ``Assuming success in the end, what will happen in near future?'' It is based on a single probabilistic inference but unfortunately neither of the two tested inference algorithms work well with it. The third control scheme is a probabilistic version of the standard nonlinear model-predictive control, which is based on optimising control signals based on a cost function. The latter two schemes are both indirect control methods and they performed comparably well in the experiments.