With advancements in hardware and network technology, the application of Virtual Reality (VR) has gradually expanded from individual immersive experiences to collaborative interactions among multiple users. Multi-user VR has shown significant application value in various fields, such as education, healthcare, industry, and entertainment. In recent years, Co-located VR has become one of the emerging and highly focused topics within the multi-user VR domain. Co-located VR enables multiple users to experience a virtual environment together within the same physical space, significantly enhancing social presence and mutual perception among VR users, making it a highly promising multi-user VR type with great potential.

　　However, the current application of Co-located VR remains constrained by the limitations of Virtual Locomotion Techniques, which are inherently bound by physical space and often lead to Spatial Desynchronization. Spatial Desynchronization refers to the inconsistency between the relative position and orientation of a user’s physical body in the real environment and that of their virtual avatar in the VR environment. This phenomenon can significantly affect users’ perception and spatial awareness of one another in VR, thereby impairing user experience and even causing safety concerns such as collisions. To cope with physical constraints and mitigate spatial desynchronization, existing Virtual Locomotion solutions for Co-located VR often sacrifice flexibility and adaptability in group structure, resulting in rigid collaborative modes and hindering the full potential of Co-located VR in multi-user collaboration scenarios.

　　While many studies have investigated Co-located VR and Virtual Locomotion Techniques across areas such as interaction design, user experience, and algorithm development, few have addressed how to effectively support Joint Action in Co-located VR—despite Joint Action being fundamental to human collaboration and social collective behavior. Existing research has primarily focused on collision avoidance during virtual locomotion, with limited attention given to how Virtual Locomotion can facilitate Joint Action and how to address the challenges posed by Spatial Desynchronization.

　　Therefore, this study aims to propose a design framework for a Virtual Locomotion system that supports Joint Action in Co-located VR, and to develop a corresponding locomotion solution tailored to collaborative scenarios. To this end, the study first analyzes and constructs the developmental process and behavioral patterns of group movement behavior associated with Joint Action, drawing from relevant theoretical foundations, in order to identify the core functionalities needed to support Joint Action. Furthermore, by investigating the environmental design of Co-located VR and conducting user studies, the research clarifies both the design elements and user experience goals of the Virtual Locomotion system. These insights are synthesized into a comprehensive design framework, which includes a coping strategy for Spatial Desynchronization centered on the mechanism of spatial recalibration. Based on this framework, an interactive prototype was designed, implemented, and subsequently evaluated through user testing.

　　The user testing of the interactive prototype verified the effectiveness of the proposed design strategy in supporting Joint Action within Co-located VR, confirming the system’s good usability and ease of use in Co-located VR environments. Finally, this study summarizes potential future optimization directions for the design strategy, providing theoretical foundations and practical suggestions to further enhance the functionality and user experience of Co-located VR systems.

　　This research provides an innovative theoretical perspective and practical design strategy for Virtual Locomotion system design for Joint Action in Co-located VR, offering valuable references for the development of Co-located VR technology applications across various industries.