Is this algorithm work for multi-layer enviroment?

[JACKLiuDay]

Hi, thank you for your brilliant work. I want to know whether DLO suitable for multi-layer structure, such as a big office building.

[kennyjchen]

Hi @JACKLiuDay -- thanks for your interest. We've seen promising results in multi-level environments (e.g., Figure 9D/E in the paper), but we haven't tested enough for a final conclusion. Empirically though, performance of DLO (and any SLAM algorithm) in multi-level buildings can depend on several variables, including whether the environment has stairs/ramps/something else, how narrow or wide they are, the FOV of your sensor, whether you tilt the sensor or not as you traverse these passageways (i.e., legged vs aerial/handcarries), etc. If you do test in these environments, we'd be interested in hearing about your experiences.

[JACKLiuDay]

tilt

Hi，thank you for reply. We are preparing to conduct a SLAM experiment in a coal mining project for robot localization in roadways. And our group will try testing with DLO, and of course other methods. We will show photos of any experimental progress in the future.