Surface-adaptive and Collision-avoiding Path Planning for Five-axis Milling

Automatic path planning for 5-axis-milling is a difficult problem if it is tried to use all degrees of freedom in an optimal way. Existing computer-based planning still require a considerable amount of interaction by experienced users, and use quite adhoc heuristics useful in practice in the automatic part. This paper presents two approaches to automatic path planning with emphasis on complex free-formed workpieces: a graph path covering approach and an iso-potential lines approach. They are based on the modelling of path-planning as optimization problems which include the behavior of the tool and the production machine with respect to kinematic aspects and collision avoidance. The algorithmic implementation of the approaches is based on discretization of the workpiece surface by a triangular mesh. A particular advantage of the iso-potentialcurve approach is the robust calculation due to the used implicit representation, and its lower computing times. The graph path covering approach is less restrictive with respect to the set of feasible solutions in which an optimum one is searched.