Simulation of Laser-Sensor Digitizing for On-Machine Part Inspection

Abstract

Integrating measurement operations for on-machine inspection in a 5axis machine tool is a complex activity requiring a significant limitation of measurement time in order not to penalize the production time. When using a laserplane sensor, time optimization must be done while keeping the quality of the acquired data. In this paper, a simulation tool is proposed to assess a given digitizing trajectory. This tool is based on the analysis of sensor configurations relatively to the geometry of the studied part.

Optimal scanning strategy for on-machine inspection with laser-plane sensor

“…The sensor used is a Kreon KZ25 laser-plane (www.kreon3d.fr). This sensor has been previously characterized [11]. Its characteritics are: a laser-line width of 25 mm for a digitizing distance d * of 50 mm from the top of FOV, a maximum view angle equal to 60 • .…”

“…Its characteritics are: a laser-line width of 25 mm for a digitizing distance d * of 50 mm from the top of FOV, a maximum view angle equal to 60 • . In the example, the trajectory is calculated with a distance between passes of I 3D = 15 mm and wth This scan path is first assessed by a simulator developed in a previous work [11], which relies on the analysis of the scanner configurations relatively to the meshed part. Green facets correspond to well digitized facets using only one configuration, whereas yellow facets belong to overlap zones ( Figure 5(b)).…”

“…The geometrical characteristics of the sensor, i.e the position of the point O c in the spindle frame are identified by the measurement of previously identified elements(elements of the table) in the working space of the machine. This yields: (10) In this measurement context, the sensor qualification is determined according to the evolution of the digitizing noise, which is assessed by scanning a reference plane for different distances and view angles ( [11]). According to this protocol carried out using the laser-plane sensor, the noise is less than 0.015 mm for the admissible scanning distances [20;50] mm from the bottom of FOV and the admissible view angles [0 • ,60 • ].…”

“…The evolution of the actual scanning noise is reported in Figure 13(a), in which the scale of colors accounts for scanning noise evolution. Scanning noise is calculated for each facet of the mesh model [11]. To do this, we create a cylinder whose basis is the triangle that defines the facet.…”

Defining Scanning Trajectory for on-Machine Inspection Using a Laser-Plane Scanner

“…The scan path planning method with overlap control ISOvScan is implemented in Matlab© . This scan path is assessed by a simulator previously developed [16]. Green facets correspond to scanned facets, whereas yellow facets belong to overlap zones (Figure 3).…”