Virtual reality-based CMM operation training and efficient probe path planning using A* and heuristic methods

Abstract

This thesis presents the design and implementation of a virtual reality (VR)-based training and path-planning system for Coordinate Measuring Machine (CMM) operations. With the increasing integration of automation and digital technologies in manufacturing, there is a growing demand for efficient, accessible, and scalable training platforms that replicate real-world inspection tasks while minimizing costs, risks, and limitations associated with physical equipment. To address this need, a virtual CMM environment is developed using Unity as a simulation engine. The system emulates CMM functionality with interactive features including the stylus qualification, coordinate system selection, dimensional measurement of geometric features, and feedback on probe movements and errors. Users gain hands-on experience by wearing a headset to learn complex metrology workflows and by navigating the environment using controllers. The thesis also introduces a path-planning method to optimize the CMM inspection routines. Traditional CMM programming methods often rely on manual sequencing, resulting in suboptimal probe travel paths and inconsistent operations, especially when dealing with intricate part geometries. To address these challenges, a hybrid planning strategy is proposed that combines intuitive heuristic guidance with the A* (A-star) search algorithm to search for the shortest collision-free path between measurement features. The method also incorporates geometric constraints, clearance verification, and support for feature-based prioritization, enabling it to adapt to a range of inspection scenarios with varying complexity. The developed system is validated through a series of simulated measurement tasks on representative 3D models, comparing the optimized path-planning results with manually defined conventional sequences. The automated planning approach achieved an 8.4% reduction in total probe travel distance compared to a manual baseline, with planning computation taking under 1 second. Qualitative and quantitative feedback from user evaluations (n = 30) conducted using the System Usability Scale further confirms the system's usability with a mean SUS score of 76.4, significantly above the 68-point benchmark, highlighting its potential as a supplementary training resource in educational and industrial contexts. By unifying immersive VR-based learning and intelligent path planning into a single platform, this research contributes to the broader objectives of smart manufacturing, digital twin development, and next-generation metrology education. The proposed solution not only enhances operator readiness but also lays the groundwork for automated and adaptive inspection strategies in manufacturing systems.