Active Visual SLAM Based on Hierarchical Reinforcement Learning
Wensong Chen,Wei Li,Andong Yang,Yu Hu,Wensong Chen,Wei Li,Andong Yang,Yu Hu
We present AVS-HRL, a modular Active visual SLAM system based on hierarchical reinforcement learning. The reward function explicitly considers the efficiency of exploration and the accuracy of mapping by utilizing the internal variables of SLAM, such as feature points distribution and loop-closure signal. Compared to end-to-end active SLAM methods, we designed a map reconstruction module that can ...
