Enhancing Path Planning Efficiency in Virtual Assembly Environments Using Fuzzy Bayesian Deep Q- Networks

Abstract

The utilization of computer virtual reality technology in virtual assembly can facilitate the design and planning of the assembly process, enhance production efficiency, and decrease economic costs. Path planning represents a critical area of development within virtual assembly, making the investigation of path planning technology essential for designing assembly paths in intricate environments. This paper introduces a deep Q-network algorithm grounded in Fuzzy Bayes, which has been evaluated within a virtual assembly context. The findings indicate that the Fuzzy Bayes-based deep Q-network algorithm demonstrates superior navigability and planning efficiency in complex, confined environments.