Automatic calculation of (dis)assembly paths for vehicle components

Overview and questions

The project is located in the field of digital assembly planning in vehicle construction. The long-term goal is the automatic generation of an assembly priority graph from which valid assembly sequences for a new vehicle can be derived. The most important basic component for an assembly priority graph is the so-called path planning: For a component in a given assembly situation, the (dis)assembly movement of the component must be calculated. The components of a vehicle are available as triangulated 3D CAD data. For small parts (e.g. screws, nuts or clips) and for rigid components (taking into account their cast-on and partially flexible fastening elements), the automated calculation of disassembly paths from the installed position in the vehicle to a body address station outside the vehicle is explored and implemented.

Procedure

The project is divided into two sub-projects. In the path planning for real (rigid) components, small parts (e.g. screws or nuts) and cast-on fastening elements (e.g. movable clips) play an important role, since components in their installed position, in the so-called close range, are always unavoidably in collision with their surrounding geometry due to them. In addition, they are often responsible for rotational movements during disassembly. The geometric algorithms we have developed for path planning in the close range take fastening elements into account during dismantling. No other academic or commercial software is currently able to do this. Other solutions simply allow some degree of penetration, which has been shown to lead to incorrect and therefore potentially invalid assembly paths, or ignore the problem completely.

For the path planning in the far distance following the planning in the close range, geometric algorithms are developed to find suitable disassembly paths to the body address stations. Here the focus is on translatory short paths with limited rotation, which guarantee a fast and easy assembly.

For the path planning in the far range following the planning in the near range, geometric algorithms are developed to find suitable disassembly paths to the body approach stations. Here, the focus is on translationally short paths with limited rotation that guarantee fast and easy assembly.

Targeted results

With the results of the project, a higher degree of automation in virtual assembly planning is achieved. The expert will receive a specialized tool that enables a fast and robust calculation of (de-)monthly paths for real components.

ManagementProf. Dr. Nicola Wolpert
Partner (external)Mercedes-Benz AG, Johannes Gutenberg University Mainz
Duration2018 - 2022
Name and position Field Email and phone Room
Professor Mathematics +49 711 8926 2697 2/368