廣義虛擬實境的發展軌跡與未來趨勢

Abstract

本研究將虛擬實境（VR）、擴增實境（AR）與混合實境（MR）統稱為廣義的虛擬實境（the broad VR），從學術與專利資料庫中檢索相關文獻，進行過去發展軌跡的探索與未來研發趨勢之預測。由學術資料庫與專利資料庫中各獲得28,566及25,134篇文獻，針對學術文獻的引用資料建立引用網絡，再對引用網絡做主路徑分析，以得到總體發展軌跡，另做集群分析與文字雲分析獲得八大研究主題分別為：VR用於外科手術訓練、VR用於暴露治療、行動AR的應用、VR用於復健、VR用於產品開發、VR用於學習、AR用於手術導航以及VR用於腦部研究。接著，將兩種文獻中所有相關論文的名稱與摘要做文字探勘，找出出現頻率較高的關鍵詞，以了解研究者較有興趣的研究議題，並得知學術研究與專利研究的時間差為0到6年。最後，統計近四年學術研究的新興主題，以指引未來專利佈局的重點方向，此結果可供產官學界在擬定廣義虛擬實境未來發展方針之參考。本研究展示一個整合式的分析步驟，成功且有效地完成廣義VR技術發展與應用的軌跡，並預測未來專利研發的方向，為有興趣做科學發展文獻回顧與技術預測的研究者提供一個新的選擇。

This study combines virtual reality (VR), augmented reality (AR), and mixed reality (MR), and names it ＂the broad VR＂. The purpose of the study is to explore the overall development trajectory and foresee the future trend of the broad VR technology. We retrieve the broad VR relevant academic literature from Web of Science and Scopus, and relevant patent literature from Derwent Innovation. By using query (＂virtual reality＂ OR ＂augmented reality＂ OR ＂mixed reality＂), we retrieve 28,536 and 25,134 articles from academic and patent databases respectively. After applying the main path analysis and cluster analysis to the citation network of the academic literature, we explore the overall development trajectory of the broad VR research and identify eight major research themes, including VR in surgery training, VR in exposure treatment, mobile AR, VR in rehabilitation, VR in learning, AR in surgery navigation, and VR in brain study. We further apply text mining to the title and abstract of all literature to identify the most mentioned terms for representing the most interesting research topics. Top 20 terms in academic literature and patent literature are identified and the time lag between them ranging from 0 to 6 years. Finally, we identify the emerging research topics in academia as the reference for future patent development. We successfully demonstrate an integrated approach for exploring the overall development trajectory of scientific literature of the broad VR technology and forecasting the future direction for patent deployment. This study provides an alternative solution for researchers who are interested in conducting a literature review and technological forecasting.