Virtual Fixtures Project 

Virtual Fixtures Project (1991 - 1994) was a pioneering mixed reality research effort at Air Force Research Laboratory (AFRL), the Stanford Center for Design Research, and NASA Ames Research Center. Conducted by Louis Rosenberg, the project enabled users to experience mixed reality for the first time by accurately registering 3D virtual objects to real-world environments in full 3D space and enabling real-time interactions between real and virtual elements. 

The system, called the Virtual Fixtures Platform, was assembled at Wright Patterson Air Force Base and employed an upper-body exoskeleton, pass-through cameras, a 3D haptics engine and 3D spatial audio to augment a user's reality with "virtual fixtures" designed to amplify performance in real-world tasks. In addition, a perceptual workstation was tested at Nasa Ames to enable users to design their own fixtures for a given task. 

Because the Virtual Fixtures Project was conducted before phrases like augmented reality, mixed reality, or extended reality had been coined, the 3D virtual objects were described as perceptual overlays (e.g., visual, auditory, kinesthetic, tactile and haptic content) added to a user's "ambient reality" to assist in real-world tasks.  

The project was successful, producing the first academic papers that show augmented reality could increase human performance in real-world tasks (Rosenberg 1992). The performance benefit of "virtual fixtures" has been validated by many third-party academic studies over the last thirty years. As a result, the VF concept is currently deployed in most robot assisted surgery systems today to support surgeons by delivering mixed reality guides, cues, and constraints. 

Another core objective of the Virtual Fixtures Project was to discover how to incorporate virtual content into real environments such that "the operator will accept the overlaid percepts as real and tangible aspects of the ambient reality." It was found that fostering a cognitive process called "distal attribution" was a critical requirement for ensuring that users perceived the real and virtual as part of a single unified reality. 

Achieving distal attribution, and thus fostering a sense of mixed reality, was found to require (i) accurate spatial and temporal registration between the real and virtual in 3D space (within human perceptual limits) and (ii) believable real-time interactivity with respect to the real and virtual such that "a natural and predictable relation exists between an operator's neuromotor activities (efference) and the subsequent changes in the sensations included in the perceptual overlay (afference)." [Rosenberg 1994].  

[ Figure Reprinted from Rosenberg - 1994 ]