Operational performance, cognitive load, visual attention, and usability of fixed-, manual-, and autonomous-camera control in single- and multiple-camera telemanipulation systems

Camera control is crucial in telemanipulation, yet its effects on human operators remain underexplored. This study examined five camera viewpoint control models in a telemanipulated cube-stacking task involving 35 participants: (1) three fixed cameras; (2) two fixed cameras plus one dynamic camera with autonomous viewpoint control; (3) two fixed cameras plus one dynamic camera with manual viewpoint control; (4) a single dynamic camera with autonomous control; and (5) a single dynamic camera with manual control. We evaluated performance (cube-stacking success rate and completion time), cognitive load (eye-tracking measures of blink rate and pupillary activity, and perceived workload), visual attention (eye-tracking measures of fixation and saccade rates), and usability. Multiple-camera models improved task success but increased cognitive load (lower blink rates and higher pupillary activity) and saccade rates. Between multiple-camera models, autonomous-camera models showed lower saccade rates. Dynamic-camera models were rated more usable than fixed cameras. These findings reveal key trade-offs in camera control design and guide the creation of more efficient, operator-friendly telemanipulation systems.