A framework for the study of six-degree-of-freedom control interfaces

This thesis investigates human behaviour when controlling Six Degree of Freedom (DoF) Interfaces. A substantial literature review forms the basis for the design of an experimental framework. An assertion is made which states that effective control interfaces will support a broad range of activity in a virtual environment. A review of motor control facilitates the design of a set of appropriate tasks and measures A series of seven experiments are presented. The series of experiments are [sic] partitioned into three studies: Object Rotation in 3DoF (three experiments); Object Manipulation in 6DoF (three experiments); Egomotion in a 3D environment (one experiment). A new rotation controller which maps 2D mouse input to 3DoF rotation is designed and implemented. It is then compared against an "integrated" 6DoF controller. The purpose of these studies is to establish an experimental paradigm that will enable designers to examine operator strategies with input devices and interfaces. From the experiments described in the studies a number of conclusions are made: (1) operator strategies cannot be identified by single measures—rather a variety of measures help disambiguate singular performance scores; (2) control strategies can be employed due to the characteristics of one interface component but can leak into behaviour with other interface components which are related in terms of the task; (3) a variety of tasks must be employed to develop a rich picture of operator behaviour with a particular interface; (4) certain characteristics of an interface can mask other performance issues when comparing interfaces; (5) travel can be analysed with a traditional tracking task; (6) the control structure of the interface must match the control structure of the task domain—if this is exceeded then in some cases performance can actually be degraded.