Abstract
In this paper, we present a method to generate surface models based on a point cloud. Aim of the model generation is to produce surface forms which are easy to locate in a robot based workcell. We obtain the point cloud from workobject CAD model or the point cloud can be generated based on actual measurements from the surface of the workobject carried out using a robot and a range sensor. In addition to presenting the different parametric forms, we estimate the uncertainties of the surface model parameters and consider the effect of uncertainties in these model parameters in workobject localization. The estimation of surface parameters and workobject localization is carried out using Bayesian -form estimation method and all noises are considered when modelling the uncertainties of the system.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the FSR2001 |
| Place of Publication | Helsinki |
| Pages | 147-152 |
| Publication status | Published - 2001 |
| MoE publication type | B3 Non-refereed article in conference proceedings |
| Event | 3rd International Conference on Field and Service Robotics, FSR 2001 - Espoo, Finland Duration: 11 Jun 2001 → 13 Jun 2001 |
Conference
| Conference | 3rd International Conference on Field and Service Robotics, FSR 2001 |
|---|---|
| Abbreviated title | FSR2001 |
| Country/Territory | Finland |
| City | Espoo |
| Period | 11/06/01 → 13/06/01 |
Keywords
- surface modelling
- uncertainty estimation
- Bayesian estimation
- Parametric surfaces
- pose estimation