A six degree of freedom (DOF) hand controller is a device that can be used for the simulta-neous control of multiple axes. Multi-axis control may be problematic as a consequence of interference i.e., the control of a certain DOF affects the simultaneous control of another. This was investigated in an experiment in which a compensatory tracking task was per-formed. In this task one DOF of a cursorn a perspective display was disturbed (externally). Subjects had to compensate this disturbance using a 6-DOF hand controller. At the same time they had to minimize input on the other (irrelevant to tracking) DOFs. It was investigated whether there were differences between tracking performance between each separate degree of freedom (X, Y, Z, Roll, Pitch, or Yaw). Furthermore, the effect of additional (irrelevant) DOFs that had to be controlled simultaneously (0, 1, or 5), was examined. With regard to the irrelevant degrees of freedom, the steering error thus was completely caused by incorrect, accidental, steering inputs. Error on the relevant DOF was a sum of this incorrect steering input and the disturbance signal. Both these errors (expressed in RMS scores) can be used to indicate the extent to which degrees of freedom interfered with each other. Through determination of the correlations between each combination of two degrees of freedom the extent to which systematic interference occurred was investigated. The experimental results show that in the 1-DOF condition tracking error was largest on Z with regard to translations and on Pitch with regard to rotations. This can be related to the effectivity of the presentation of the z-axis (i.e., used depth cues and compression) in the used perspective display. Furthermore, performance on relevant as well as irrelevant DOFs decreased when the number of visible degrees of freedom that had to be controlled increased. These limitation are attributed to the limited information processing capacity of the human operator. In relation to the other DOFs, this performance decrement for Z was substantially larger whereas it was smaller for X. Again this may be related to the effectivity of presenta-tion of the different axes on the display. Input on a relevant DOF and input on an irrelevant DOF were always significantly corre-lated. The amount of cross-talk between degrees of freedom did not change with the number of DOFs that h had to be controlled. For half the combinations cross-talk even remained the same in the conditions without any visual information on the irrelevant DOFs. Therefore, it seems that cross-talk mainly results from motor limitations of the operator. Increasing the amount of haptic information in the hand controller, probably will help the operator to distinguish the degrees of freedom more easily. This may result in a reduction of cross-talk and better control.
Selectieve besturing van vrijheidsgraden met een 3D muis wordt bepaald door de kwaliteit van de relevante virtuele informatie in combinatie met de hoeveelheid teruggekoppelde informatie over niet-relevante vrijheidsgraden. Systematische overspraak wordt voornamelijk bepaald door motorische beperkingen.