Non ti piace? Non importa! Puoi restituircelo entro 30 giorni
Non puoi sbagliarti con un buono regalo. Con il buono regalo, il destinatario può scegliere qualsiasi prodotto della nostra offerta.
30 giorni per il reso
This book is concerned with planning and acting under§uncertainty in§partially-observable continuous domains. It focusses§on the problem§of mobile robot navigation given a known map. The§dominant paradigm§for robot localisation is to use Bayesian estimation§to maintain a§probability distribution over possible robot poses. §In contrast,§control algorithms often base their decisions on the§assumption that§the most likely state is correct, rather than§considering the entire§distribution.§§This book formulates an approach to planning in the§space of§continuous parameterised approximations to§probability distributions.§Theoretical and practical results are presented which§show that, when§compared with similar methods from the literature,§this approach is§capable of scaling to larger and more realistic problems.§§The algorithms have been implemented and demonstrated§during real-time§control of a mobile robot in a challenging navigation§task. Results§show that this approach produces significantly more§robust behaviour§when compared with heuristic planners which consider§only the most§likely states and outcomes.