information for better poses configuration estimation. So in
these works, anchor n odes [9], weak links [1,6] are introduced
to solve the problem. The vision based SLAM in low dynamic
environment has also been studied. In Ref. [10], multiple poses
formed a view cluster, in which the images with the similar
view would be updated over time. This method can tell
whether a frame is out-of-dated but cannot show which part
has been changed as it was a sparse visual feature based
method.
Most existing methods dealing with SLAM in dynamic
environment is based on 2D laser SLAM. In Refs. [11,12], the
set of scans in global coordinates was updated by sampling
after each new session to build an in-dated map. In their work,
poses were estimated by SLAM at the first session. For the
later sessions, the poses were estimated by localization, not
included in the SLAM framework. In Refs. [8,13], both works
described the dynamic with each cell in grid occupancy map
having an inde pendent Markov Model. In Ref. [7], a dynamic
environment map was modeled as a pose graph. After each
session, the out-of-dated poses are identified and removed
based on 2D occupancy grid map built from the laser data. In
Ref. [14], the poses related to the low dynamics were removed
to enhance the robust of the optimizer.
In the context of RGBD SLAM, most works apply the
graph mode l, followed by a global optimization backe nd. In
Ref. [15], both visual features and depth information are
employed to form an edge in the pose graph. Besides the
formulation, an environment measurement model was pro-
posed for pose graph edge selection in Ref. [3]. In Ref. [4],a
dense visual odometry is used as frontend to formulate the
pose graph, which is more accurate than sparse featur e based
visual odometry. In Ref. [5], non-rigid deformation is com-
bined with the pose graph optimization for globally consistent
dense map, which takes the map mesh into consideration.
Extension of these RGBD SLAM systems to multi-session can
be achieved by applying the methods developed in Refs.
[1,6,9]. But the detection of dynamics by simply using the
laser based method is difficult, as the methods employed an
occupancy grid map for information fusion and de-noise.
When it comes to the case of RGBD sensor, the 3D o ccu-
pancy grid map is intractable due to the high complexity. So
methods should be developed on the raw senso r data, making
the problem more challenging.
Besides the mechanism for dealing with dynamic envi-
ronment, a framework for RGBD SLAM also needs node
pruning to keep the computational complexity noncumulative.
Fig. 1. A comparison of the reconstructed low dynamic environment in point cloud with 10 sessions mapping using multi-session SLAM without considering the
low dynamics (top) and proposed framework considering the low dynamics (bottom). One can see that the book, box, bottles and plastic bags are repeated, making
the scene with incorrect duplicated information. The book and the chip can are highlighted using light and dark orange rectangles. Their out-of-dated positions are
highlighted using red rectangles. The arrows demonstrate the correspondence.
92 Y. Wang et al. / CAAI Transactions on Intelligence Technology 1 (2016) 90e103