Cooperative haptic manipulation of virtual environments

虚拟环境的协同触觉操作指的是在多个（可能是远程）用户共同与虚拟物体交互时，为他们的双手提供物理精确的力感。这种操作在基于虚拟环境的训练应用程序中是有益的，这些应用程序需要多个用户的灵巧操作，包括许多外科手术。协作触觉可以消除当前单点触觉的严重灵巧性和感知局限性。在外科手术训练中，合作触觉可以：(i)为涉及双手的手术（如缝合）提供经验；（ii）使专家外科医生能够在现场或远程指导具有不同经验水平的新手外科医生；（iii）允许几位外科医生合作实践和/或计划针对患者特定解剖的具有挑战性的手术。目前具有触觉反馈的模拟器只能单手、单用户、点状操作相对简单的虚拟环境，包括可变形或刚性物体。此外，现有的实现通常包括特定于仿真的触觉渲染算法。因此，它们为有限数量的交互提供了强制反馈，并且作为培训工具的影响有限。提出的研究重点是为多个可能的远程用户提供物理精确的力，同时他们一起操作包括软硬组织在内的复杂虚拟解剖结构，并通过各种建模形式进行模拟。为此，工具与骨骼、组织、肌肉和韧带之间相互作用的局部模型将提供给每个触觉设备。这些局部模型将通过控制与全局模拟同步。面对复杂模拟或远程交互造成的延迟，预测将提高局部模型的准确性。在这项工作中开发的建模、仿真和控制方法也将用于远程机器人（远程外科）和虚拟现实的其他领域，如协作设计/原型和娱乐。

Cooperative haptic manipulation of virtual environments refers to providing several, possibly remote users with physically-accurate force sensations in both hands when they together interact with virtual objects. Such manipulations are beneficial in virtual environment-based training applications that require dexterous manipulations by multiple users, including many surgical procedures. Cooperative haptics can remove the severe dexterity and perceptual limitations of current single point haptics. In surgical training, cooperative haptics can: (i) provide experience for procedures that involve both hands, like suturing; (ii) enable expert surgeons to mentor novice surgeons with various levels of experience, on site or from a remote location; and (iii) allow several surgeons to cooperatively practice and/or plan challenging procedures on patient-specific anatomy. Current simulators with haptic feedback enable only single handed, single user, point-like manipulations of relatively simple virtual environments comprising either deformable or rigid objects. Furthermore, existing implementations typically include simulation-specific haptic rendering algorithms. Therefore, they provide force feedback for a restricted number of interactions and have limited impact as training tools. The proposed research focuses on providing physically-accurate forces to multiple, possibly remote users while they together manipulate complex virtual anatomy comprising soft and hard tissues and simulated via a variety of modeling formalisms. To this end, a local model of the interaction between the tool and bone, tissue, muscle, and ligaments will be supplied to each haptic device. These local models will be synchronized with the global simulation via control. Prediction will improve the local model accuracy in the face of delays due to complex simulations or to remote interaction. The modeling, simulation, and control approaches developed during this work will also be of use in telerobotics (telesurgery) and in other areas of virtual reality, like collaborative design/prototyping and entertainment.