Photogrammetric Road Reconstruction for Vehicle Development
Publication from Digital
Gerhard Paar, W. Reinalter
7th ISPRS Conference on Optical 3-D Measurement Techniques, October 3-5, 2005, Vienna, Austria. , 1/2015
Uneven road surfaces represent a major excitation source for any vehicle because of their significant influence particularly on the load spectrum, the required space and the ride comfort of the vehicle. Reliable simulation models that provide a high confidence level in an early phase of the product development process are necessary in order to reduce the development risk. To carry out these simulations, a sufficient digital representation of the relevant road track surface sections is required. This paper describes the photogrammetric refinement of an existing low-resolution geo-referenced surface digitization on two of the MAGNA STEYR test areas in Graz “Belgian Block” and
“rough road”. High resolution imaging without geo-reference uses a prosumer camera on a mobile platform, triggered with the vehicle – internal odometer. Image matching and photogrammetric adjustment lead to locally consistent orientations and a high resolution digital elevation model (DEM) in a dedicated road co-ordinate system. Structural DEM matching between the geo-referenced low-resolution DEM and the unoriented high-resolution DEM enables a geo-referenced refinement regardless of image texture. The systematic error between the high – resolution and the low – resolution reconstruction is below 2 mm. Global mean elevation error is below 4 mm, elevation resolution is
around 1 mm. An extensive analysis shows the usability of the road data for the vehicle development process at MAGNA STEYR. The refinement method based on structural DEM matching is suited for low-cost resolution
improvement of existing surface reconstructions.