A Hidden Markov Model based efficient next best viewpoint planning for autonomous exploration

摘要: A next best viewpoint planning algorithm based on Hidden Markov Model (HMM) was presented for autonomous exploration. The grid map and probabilistic assessment model HMM established to evaluate the exploration perspective. Leveraging relative information entropy, maximum gain of candidate viewpoints were assessed. After establishment actual scene graph rule, state transfer matrix observation occupied transition derived. Used Bayesian filtering update diagram, path selected from starting point return point. By simulation experiment different complexity, efficiency its good robustness confirmed.