Paper · arXiv 2308.16246
Active Neural Mapping
Zike Yan, Haoxiang Yang, Hongbin Zha
The paper introduces Active Neural Mapping, a method using a coordinate-based implicit neural representation to guide an agent for active exploration and mapping in unseen environments, leveraging the uncertainty in the neural field weights and geometric information.
Abstract
We address the problem of active mapping with a continually-learned neural scene representation, namely Active Neural Mapping. The key lies in actively finding the target space to be explored with efficient agent movement, thus minimizing the map uncertainty on-the-fly within a previously unseen environment. In this paper, we examine the weight space of the continually-learned neural field, and show empirically that the neural variability, the prediction robustness against random weight perturbation, can be directly utilized to measure the instant uncertainty of the neural map. Together with the continuous geometric information inherited in the neural map, the agent can be guided to find a traversable path to gradually gain knowledge of the environment. We present for the first time an active mapping system with a coordinate-based implicit neural representation for online scene reconstruction. Experiments in the visually-realistic Gibson and Matterport3D environment demonstrate the efficacy of the proposed method.