In addition, other algorithms and models for the shortest path planning problem have also been proposed. Similar researches that applied TSP algorithms can be found in. , used the TSP algorithm based on the annealing method to optimize the inspection sequence. proposed an improved 3D ACO algorithm that can quickly obtain the shortest inspection path. applied ACO algorithms to optimize the inspection path for prismatic parts. developed a path planning method for aerospace structures based on the adjacent graph, LKH algorithms were used to minimize the length of the inspection path. Due to the similarity to the traveling salesman problem (TSP), TSP heuristic-based algorithms, such as Ant colony algorithm (ACO), Genetic algorithm and Lin-Kernighan algorithm (LKH), have been used to plan the shortest path. Many researches focused on finding the shortest inspection path that travels all inspection points in sequence and only once. Therefore, the objective of this paper is to optimize both the number of probe orientations (NPOs) and the inspection path length when planning an inspection path. Third, the workload of calibration and compensation may increase since probe calibration is needed for each probe orientation before the inspection execution. Second, FOC may increase the probe temperature and thus affects the inspection accuracy. First, frequent orientation changes (FOC) can lead to rotation and clearance errors of the rotating axis. After that, these batches are inspected in sequence. However, in practice, when a part with slopes or declined holes is inspected, the inspection points are usually divided into batches with respect to probe orientations. Previous methods often optimized a collision-free inspection path to reduce the inspection time cost.
Inspection path planning methods are of great significance to OMI systems, which can considerably increase inspection efficiency. MOA can effectively decrease the NPOs compared with SPA, and the path length can be even shorter than that of SPA in some complicated cases.Ī 5-axis on-machine inspection (OMI) system usually installs a touch trigger probe on a 5-axis CNC machine to inspect part quality after or during the manufacturing process.
Simulation experiments were conducted on three parts, the results show that both MOA and SPA can obtain a collision-free inspection path. To avoid possible collisions during the inspection process, a safe box that contains the inspection part was defined before planning the path, which enables the probe to cross obstacles by moving to the safe planes of the safe box.
SPA firstly uses the TSP algorithm to obtain the shortest inspection sequence, then minimizes the NPOs by finding orientations that can inspect most consecutive points along the planned inspection sequence according to the O-P relation schema. The traveling salesman problem (TSP) algorithm was used to plan the inspection path for each group's points, finally the total path was obtained by connecting paths of different groups in sequence. MOA was designed to find the minimum orientations that can measure all the inspection points according to the O-P relation schema, then the inspection points were categorized into groups by the selected orientations. Then based on the O-P relation schema, two algorithms – the minimum orientation algorithm (MOA) and the shortest path algorithm (SPA), which consider NPOs and path length in different priorities were proposed. To optimize the NPOs and the path length, this paper firstly proposes the concept of orientation-point (O-P) relation schema which includes all valid O-P inspection cases. However, the number of probe orientations (NPOs) also has a great influence on inspection efficiency and accuracy, especially for parts with complicated features. Most researches only optimize the inspection path length. Inspection path planning methods are of great importance to OMI systems.
0 kommentar(er)
