Active Pose SLAM

Abstract

Abstract-We present an active exploration strategy that complements Pose SLAM [1] and optimal navigation in Pose SLAM [2]. The method evaluates the utility of exploratory and place revisiting sequences and chooses the one that minimizes overall map and path entropies. The technique considers trajectories of similar path length taking marginal pose uncertainties into account. An advantage of the proposed strategy with respect to competing approaches is that to evaluate information gain over the map, only a very coarse prior map estimate needs to be computed. Its coarseness is independent and does not jeopardize the Pose SLAM estimate. Moreover, a replanning scheme is devised to detect significant localization improvement during path execution. The approach is tested in simulations in a common publicly available dataset comparing favorably against frontier based exploration.

A high-efficiency, information-based exploration path planning method for active simultaneous localization and mapping

“…The exploration in an unknown environment to active SLAM have been studied by many researchers. 11,[14][15][16][17]25,26,29,[31][32][33][34][35][36][37][38][39][40][41][42] We will give a brief review of these methods as the following three categories.…”

Adaptive sampling of transient environmental phenomena with autonomous mobile platforms

“…From these principles, entropy was defined as: Entropy, conditional entropy, and mutual information have been used extensively in SLAM techniques [74][75][76][77], sensor placement [35,78,79], optimal navigation [80], and robotic sampling [15,81].…”

Autonomous Exploration Using an Information Gain Metric

“…Other work 9,10 has detailed an approach to exploration that uses entropy metrics for path selection in addition to revisiting actions to improve localization. Our technique is most similar to this work; the key difference being that we use the iSAM2 nonlinear optimization engine to efficiently compute the effect of potential loop closures, which are discovered by executing kinematically feasible trajectories, while the previous work 9,10 used an approximate sparse information filter to compute the value of the information, which can be added through loop closure while following a probabilistic road map.…”

“…Our technique is most similar to this work; the key difference being that we use the iSAM2 nonlinear optimization engine to efficiently compute the effect of potential loop closures, which are discovered by executing kinematically feasible trajectories, while the previous work 9,10 used an approximate sparse information filter to compute the value of the information, which can be added through loop closure while following a probabilistic road map. It is not clear from Valencia et al 9 if their approach is scalable to areas larger than a few rooms, whereas our approach has been evaluated on medium-sized floor plans in this work, and our mapping system on large buildings in Rogers et al 11 Here, we document a method of autonomous exploration that uses an information gain metric and SLAM techniques to map an unknown area. We show that this technique improves the efficiency and accuracy of existing techniques and that it can be implemented on widely available robot hardware.…”