Roy Ruddle: Virtual Humans

Project title: Case studies of direct interaction with virtual humans in immersive design environments

Investigators: Dan Amos, Prof Dylan Jones, Proj D-T Pham, Dr Roy Ruddle, and Justin Savage

Funding: EPSRC (1998-2001)

Note: this research took place while Roy Ruddle was employed at Cardiff University

This project studied ways of interacting directly with virtual humans within the context of the types of task performed during the ergonomic design of an automotive factory. The project's research took the form of a series of case studies that investigated the following:

Movement algorithms for cluttered virtual worlds
Object manipulation in real and virtual environments
Manipulating bulky virtual objects
"Heavy" virtual objects
Data on VE sickness were also gathered throughout the project

Movement algorithms for cluttered virtual worlds

We studied the effectiveness of view-direction, body-direction, and independent movement interfaces using a task in which participants searched small (room-sized) but cluttered environments for targets that were placed in a small number of marked possible locations. Contrary to expectations it was view-direction movement that proved the most efficient. However, the most important finding was the great difficulty that participants had completing the task with all the types of movement. Participants found it very difficult to remember where they had searched, even though they could see the whole environment by standing in one place and looking around. In fact, on 20% of trials participants ended up revisiting at least half of the environment while trying to complete the task.

A view inside one of the cluttered VEs. Visible are two sides of the boundary wall (coloured orange and magenta), target/distracter cylinders(blue-topped) and other cylinders (green only), and a target (the white square). The horizontal field of view is 48 degrees.

Widening the field of view to 103 degrees (compared with 45 degrees) produced a three-fold reduction in the number of inefficient searches but did not eliminate the problem. Further invesigations are taking place in a follow-on project understanding disorientation in cluttered virtual environments.

As well as the paper below, results of this research were presented at the:

International Conference on Spatial Cognition (EPCE 2000)
EURESCO Conference on Modelling for Wayfinding Services, 2003 (invited talk)

Publications

Ruddle, R. A., & Jones, D. M. (2001). Movement in cluttered virtual environments. Presence: Teleoperators and Virtual Environments, 10, 511-524. Video.

Object manipulation in real and virtual environments

The focus of this research was to compare manual rotation of virtual objects in real and virtual environments. We adopted the mental rotation paradigm and made some of the very first studies of freeform manual rotation in the real world, along the way developing metrics that characterised participants' rotational behaviour. This was followed by a series of virtual object rotation that showed virtual rotation behaviour was broadly similar to real-world rotation (as predicted) but substantially less efficient. Some of these efficiencies lie in the difficulty participants had perceiving the orientation of the virtual objects. Other inefficiencies relate to the execution of motor movements to change a virtual object's orientation, caused in part by a haptic inconsistency between the virtual object and interface prop.

Publications

Ruddle, R. A., Huddart, S. A., & Jones, D. M. (1999). Interaction in immersive virtual environments: Rotating objects with an instrumented prop. Proceedings of the Human Factors and Ergonomics Society 43rd Annual Meeting (HFES'99), 1214-1218.

Ruddle, R. A., & Jones, D. M. (2001). Manual and virtual rotation of a three-dimensional object. Journal of Experimental Psychology: Applied, 7, 286-296.

Manipulating bulky virtual objects

This research was divided into two parts. First, studies were made of the effect of incorporating low-level control intelligence into VE interface software, evaluated using a task in which participants collaborated with an autonomous virtual human to carry an object from one place to another, sometimes having to avoid obstacles en-route (video). The studies showed that automating the low-level aspects of control produced a three-fold reduction in the time it took participants to complete the task. The studies also demonstrated the benefits of providing participants with a tethered rather than a human's-eye view, even when a head-mounted display is being used.


Sticky-small interface (Experiment 1)	Human's-eye view (Experiment 2)	Tethered view (Experiment 3)

The second part of the bulky objects research used the piano movers' problem as a paradigm for investigating the effect on manipulation performance and behaviour of: (a) flexible rules of interaction (video), and (b) symmetric and asymmetric action integration for cooperative manipulation (video). Flexible interaction produced a saving of more than 50% in the time it took participants to perform the task. The collaboration overhead was negligible for the simplest tasks that were studied, but approximately 50% for complex tasks. In common with other studies involving collaborative VEs, participants performed a huge amount of verbal communication to try and compensate for sensory defficiencies in the interface (participants could see each others' virtual self but not feel each others' actions). Our analysis of this communication involved transcribing, categorising and analysing more than 50,000 words from 12 hours of video!


An individual performing the piano movers' problem	Cooperative version of the same task. One person sees the top view, controlling the virtual human that is wearing the blue hat, while the other person sees the bottom view and controls the virtual human wearing the red hat.

Publications

Ruddle, R. A., Savage, J. C., & Jones, D. M. (2002). Evaluating rules of interaction for object manipulation in cluttered virtual environments. Presence: Teleoperators and Virtual Environments, 11, 591-609.

Ruddle, R. A., Savage, J. C., & Jones, D. M. (2002). Implementing flexible rules of interaction for object manipulation in cluttered virtual environments. Proceedings of the ACM Symposium on Virtual Reality Software and Technology (VRST'02), 89-96.

Ruddle, R. A., Savage, J. C., & Jones, D. M. (2002). Symmetric and asymmetric action integration during cooperative object manipulation in virtual environments. ACM Transactions on Computer-Human Interaction, 9, 285-308.

Ruddle, R. A., Savage, J. C., & Jones, D. M. (2002). Symmetric and asymmetric action integration during cooperative object manipulation in virtual environments. Interactions, 9(6), 9-10.

Ruddle, R. A., Savage, J. C., & Jones, D. M. (2002). Verbal communication during cooperative object manipulation. Proceedings of the ACM Conference on Collaborative Virtual Environments (CVE'02), 120-127.

Ruddle, R. A., Savage, J. C., & Jones, D. M. (2003). Levels of control during a collaborative carrying task. Presence: Teleoperators and Virtual Environments, 12, 140-155.

"Heavy" virtual objects

Providing feedback to help participants: (a) lift virtual objects at the same rate as in the real world, and (b) perceive the forces that heavy virtual objects exert on the human body. The findings from this research were published in Dan Amos' PhD thesis Immersive virtual environments to aid manual lifting simulations.

VE sickness

A meta analysis of all the sickness data was performed, highlighting the impact of looking steeply downwards at one's virtual feet on the severity of symptoms that participants sufffered.

Publications

Ruddle, R. A. (2004). The Effect of Environment Characteristics and User Interaction on Levels of Virtual Environment Sickness. Proceedings of IEEE Virtual Reality (VR'04), 141-148. Video.