Automatic data driven vegetation modeling for lidar simulation

Jean-Emmanuel Deschaud (1) , David Prasser (2) , Freddie Dias (2) , Brett Browning (2) , Peter Rander (2)

1 CAOR - Centre de Robotique
2 NREC - National Robotics Engineering Center [Carnegie Mellon]

Traditional lidar simulations render surface mod-els to generate simulated range data. For objects with well-defined surfaces, this approach works well, and traditional 3D scene reconstruction algorithms can be employed to au-tomatically generate the surface models. This approach breaks down, though, for many trees, tall grasses, and other objects with fine-scale geometry: surface models do not easily represent the geometry, and automated reconstruction from real data is difficult. In this paper, we introduce a new stochastic volumetric model that better captures the complexities of real lidar data of vegetation and is far better suited for automatic modeling of scenes from field collected lidar data. We also introduce several methods for automatic modeling and for simulating lidar data utilizing the new model. To measure the performance of the stochastic simulation we use histogram comparison metrics to quantify the differences between data produced by the real and simulated lidar. We evaluate our approach on a range of real world datasets and show improved fidelity for simulating geo-specific outdoor, vegetation scenes.

Domaines: Robotique [cs.RO]; Vision par ordinateur et reconnaissance de formes [cs.CV]; Modélisation et simulation

Jean-Emmanuel Deschaud, David Prasser, Freddie Dias, Brett Browning, Peter Rander. Automatic data driven vegetation modeling for lidar simulation. ICRA, May 2012, St. Paul, United States. pp.5030 - 5036, ⟨10.1109/ICRA.2012.6225269⟩. ⟨hal-01097398⟩