Automatic recognition of machining features from a solid model using the 2D feature pattern

Abstract

Feature recognition systems are now widely identified as a cornerstone for conceiving an automated process planning system. Various techniques have been reported in the literature, but a few of them acquired a status of generic methodology. A flexible and robust approach is demanded for recognising a wide variety of features, e.g., non-interacting, interacting circular and slanting features. This research aims to exploit the concept of the ray -firing technique, in which a 2D surface pattern for each feature is generated and information is extracted from these patterns to correlate it with the corresponding machining features. The system first defines a virtual surface and then probing rays are dropped from each point of this surface to the 2.5D features of the B-rep solid model. According to the length of rays between the bottom face of the 2.5D machining features and the virtual surface, 2D feature patterns are formed for each machining feature. Finally, features are recognised using an algorithm described in this article. Different types of examples have been considered to demonstrate the effectiveness of the proposed approach.

CAD integrated automatic recognition of weld paths

“…Furthermore, publicly available industrial implementations and commercially available feature recognition systems exist [33]. This work has focused on adapting feature recognition techniques from the works outlined in Section 2 such as [1,18,28] into welding applications in the CAD environment. This has led to restrictions in comparing the method to existing approaches, with the exception of the welding joint tool in NX.…”

“…These 2D cross-sections can then be used as subject to further analysis by applying ray casting methods from the center of the weld to determine a form for point visibility. This is essentially the same as the methods [1,18,28], however, applied to a sliced 2D section for welds. Based on this point visibility, measured as a range of consecutive open space around the weld, the assumption is that a sense of weldability, or accessibility can be determined.…”

“…It is being sliced to recognize 2D cross-section/features. The importance of feature recognition systems was presented in [1]. The paper reports on a method of feature recognition using light rays from a given direction, i.e., machining direction.…”

“…This captured knowledge, or logic, can be redeployed at a given time to automate decisions and design parameters, and provide smarter and more optimized solutions taking available information into account. However, in order to build such systems, methods for recognizing of specific features are considered a necessity [1,2].…”