SOLO-SLAM: A Parallel Semantic SLAM Algorithm for Dynamic Scenes

. 2022 Sep 15;22(18):6977. doi: 10.3390/s22186977

Algorithm 1 Update dynamic probability and semantic properties of map points

\begin{array}{l} Input : M^{in_view} K F^{o p t i m i z e} \\ Output : M^{p r o j e c t} R^{s e m} \\ For k f^{o p t i m i z e} i n K F^{o p t i m i z e} do : \\ / / Get instance segmentation results based on ROS \\ R^{s e m} \leftarrow R e q u e s t I n s t a n c e S e g m e n t a t i o n (k f^{o p t i m i z e}); \\ r^{d y n a m i c} \leftarrow G e t D y n a m i c R e g i o n (R^{s e m}); \\ r^{s t a t i c} \leftarrow G e t S t a t i c R e g i o n (R^{s e m}); \\ M^{p r o j e c t} \leftarrow G e t P r o j e c t i o n o f M a p p o i n t s (M^{in_view}); \\ / / Update only the semantic attributes of map points \\ / / whose projections lie in static regions \\ For m^{p r o j e c t} i n M^{p r o j e c t} do : \\ m^{d y n a m i c} \leftarrow U p d a t e D y n a m i c P r o b a b i l i t y (m^{p r o j e c t}, r^{d y n a m i c}); \\ If (m^{p r o j e c t} in r^{s t a t i c}) then \\ m^{s e m} \leftarrow U p d a t e S e m a n t i c P r o p e r t y (m^{p r o j e c t}, r^{s t a t i c}); \\ End If \\ End For \\ End For \end{array}