Input for three dimensional navigation using two joysticks

Details Input for three dimensional navigation using two joysticks Input for three dimensional navigation using two joysticks

1997-11-30

2001-02-06

Shankar, Vijay (Department: 2778)

Computer graphics processing and selective visual display system

Display peripheral interface input device

Cursor mark position control device

C345S156000, C345S157000

Reexamination Certificate

active

06184867

ABSTRACT:

BACKGROUND
1. Field of the Invention
The invention relates to information processing and multi-media systems, and in particular to a two-joystick, bulldozer-like interface for navigating in a virtual three-dimensional space.
2. Background of the Invention
Some Preliminary Definitions
FIG. 1
is a graph illustrating a 3D notation convention used in this patent application. The convention is designated generally by the numeral
10
and includes a left-right horizontal X axis
12
, a vertical Y axis
14
, a point-of-view Z axis
16
, and rotation
18
about the Y axis is designated by the symbol &THgr;. The X axis
12
is from the viewer's left side to the right side. The Y axis
14
is from the bottom toward the top, and the Z axis
16
is in the direction the viewer is facing. Pitch is rotation about the X axis, yaw is rotation about the Y axis, designated by the symbol &THgr;, and roll is rotation about the Z axis.
Isometric devices have constant location and infinite resistance to displacement. Isotonic devices have constant resistance (e.g., zero) to displacement. Elastic devices have resistance force proportional to displacement.
We live in a three-dimensional (3D) world. Technological advances in computer graphics hardware, software and display systems will soon make real time 3D capabilities available to all mainstream computer systems. Furthermore, 3D technology is also beginning to be integrated into the Internet technology, making it possible for designers and users on various platforms to escape from the flatland of 2D HTML pages to share 3D information across the world. Virtual Reality Modeling Language (VRML), a language for modeling three-dimensional scenes and scripting animation on the World-Wide-Web (WWW), is an emerging standard for such a purpose. Using VRML, web designers can construct 3D “worlds” in which a remote user can navigate.
However, disorientation and difficulty of knowing how to move in current 3D interfaces are intolerably inhibiting the widespread applications of 3D technologies. Designing effective navigation interfaces is both a difficult challenge and an opportunity to the human factors community. First, it is a rare opportunity. As if we could move back in time to have the chance to design interfaces for aircraft and motor vehicles, we are now at the beginning of a new era when it is not (yet) too late. Furthermore, the “virtual navigation craft” does not impose on us any mechanic or dynamic constraint as in aircraft and motor vehicles. The only fundamental constraints to virtual 3D navigation interface lies in accommodating human capabilities and limitations. In other words, designing virtual world navigation interfaces is less like designing motor vehicles but more like searching for the “magic carpet” in the virtual world. A magic carpet takes us where we want to go without our having to worry about the mechanical details of getting there.
Although few fundamental constraints exist to designing virtual 3D navigation interfaces, there are many practical requirements for a navigation interface to gain rapid acceptance. Particularly, we have to consider the following:
integration with the existing GUI interfaces and tasks. Like any new interface technology, 3D navigation is in a bootstrap situation: not enough application and content may make it unjustified to have a special interaction device and lack of a usable interface and user population make developing 3D content less attractive. The solution lies in seamless integration of the 3D navigation devices with the existing GUI interfaces;
low cost, for the same reason as above; and
both novice and expert “friendly”. An acceptable navigation interface should be easy to learn in a few minutes while offering high performance to expert users.
Although there are many VRML browsers available today, the user interfaces of these browsers are very similar to each other. The basic interface design was set in Silicon Graphics' WebSpace™ (later known as CosmoPlayer™), the first commercially available VRML browser, which in turn was based on the 3D user interface work previous to VRML, see Zeleznik, R. C. et al., An Interactive 3D Toolkit For Constructing 3D Widgets, Proc. ACM SIGGRAPH '93, pp. 81-84, 1993.
With these existing browsers, 3D navigation is done by mapping 2 degrees of freedom (DOF) mouse cursor movements onto various translation and rotation degrees of freedom, according to the selected mode. For example, the WorldView™ browser, see WorldView™ VRML 2.0 Browser, Intervista Inc., at “http://www.intervista.com”, has the following modes: walk (Z translation and Y rotation in rate control), pan (X and Y translation in rate control: FIG.
2
), turn (X and Y rotation in rate control), roll (Z rotation in rate control), study (X and Y rotation of the world, instead of self, in position control) and goto (FIG.
3
). A user can switch between these different mapping schemes by clicking on appropriate buttons. Some browsers also provide a restore mode to readjust the Y coordinate of the view coordinates with that of the world coordinates. Much effort had been spent on fine-tuning the mapping transfer functions in various modes.
When navigating in VRML worlds with the current browsers, users typically find themselves off their targets, facing upside-down, or lose their locations in the world. Multiple factors, such as low frame rate that delays the control feedback and sometimes badly designed “cueless” worlds themselves contribute to the “lost in virtual world” problem. One of the most critical factors lies in the mouse mapping navigation technique itself. Among the many drawbacks of this technique, the following are the most noticeable:
Mode switching. It is known that mode switching in user interface in general should be avoided.
Mode switching causes inconsistent response to the same input. For the same mouse movement, the results are different depending on the current mode. It is well known that when consistent mapping exists, human information processing behavior tends to become an “automatic process” which requires little central capacity, attention, or effort. In contrast, when consistent mapping is absent, human behavior tends to be a “controlled processes” which requires effort, attentional resource, and central capacity.
In many of these modes, cursor motions are mapped to movement “speed”. The farther one moves the cursor from the initial click position, the faster the movement is. In other words, the cursor displacement is used for rate control, which is well suited for navigation where smooth and controllable speed is desirable. Experiments have shown that effective rate control requires self-centering mechanisms in devices such as isometric or elastic joysticks. Isotonic devices such as the mouse are poor in rate control tasks. Note that when one uses a rate controlled joystick (such as the TrackPoint™ by IBM) to move the cursor which is in turn mapped onto movement speed, the self-centering effect in the joystick is not utilized, since the self-centering variable (e.g., force) is not directly mapped onto the movement speed.
Visual information provided by the mouse pointer is counter-intuitive. Typically, when users want to approach an object in the world, they tend to bring the mouse pointer onto the object. However, due to the mapping function of the mouse (both speed and direction), such an action may cause unexpected results in which users tend to overshoot position and orientation.
Clearly a need exists for different devices and techniques to replace the status-quo interface. One apparent choice is the six degree-of-freedom hand controllers such as the “Spaceball” (see Zhai, S., Human Performance in Six Degree-of-Freedom Input Control, Ph.D. Thesis, University of Toronto, for a review of six degree-of-freedom devices, available at “http://vered.rose.toronto.edu/people/shumin_dir/publications.html”).
Several reasons make this option difficult. First, these devices have been relatively expensive due to the small market size and they are not integrated with the general

Affiliated with

Also associated with

Rating

Input for three dimensional navigation using two joysticks does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Input for three dimensional navigation using two joysticks, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Input for three dimensional navigation using two joysticks will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2586600