Training a next-best-view (NBV) planner for active cross-domain self-localization is an important and challenging problem. Unlike typical in-domain settings, the planner can no longer assume the environment state being constant, but must treat it as a high-dimensional component of the state variable. This study is motivated by the ability of recent visual place recognition (VPR) techniques to recognize such a high-dimensional environment state in the presence of domain-shifts. Thus, we wish to transfer the state recognition ability from VPR to NBV. However, such a VPR-to-NBV knowledge transfer is a non-trivial issue for which no known solution exists. Here, we propose to use a reciprocal rank feature, derived from the field of transfer learning, as the dark knowledge to transfer. Specifically, our approach is based on the following two observations: (1) The environment state can be compactly represented by a local map descriptor, which is compatible with typical input formats (e.g., image, point cloud, graph) of VPR systems, and (2) An arbitrary VPR system (e.g., Bayes filter, image retrieval, deep neural network) can be modeled as a ranking function. Experiments with nearest neighbor Qlearning (NNQL) show that our approach can obtain a practical NBV planner even under severe domain-shifts.
Active Map-Matching: Active Self-localization by VPR-to-NBV Knowledge Transfer
