Design and Evaluation of a Touchscreen-Based Teleoperation Interface for Robotic Manipulators

As robotic manipulators become increasingly vital for hazardous tasks like nuclear decontamination, the interface used to control them becomes a critical bottleneck. Conventional joysticks often require high cognitive effort and extensive training to manage the simultaneous demands of path tracking and force regulation. This study investigates whether a touchscreen-based interface, which maps finger movements directly to robotic motion, can provide a more intuitive and efficient alternative for surface-contact tasks.

The researchers developed a touchscreen interface that integrates visual feedback with direct positional control. To evaluate its effectiveness, they conducted a comparative study with 20 participants who performed surface-swabbing tasks using a Franka Emika Panda arm. The robot was located in Italy, while participants operated it from the UK, introducing real-world network latency. Participants completed tasks under three conditions: a standard joystick, the new touchscreen interface, and a single-click autonomous mode. The team measured performance through task completion time, path accuracy, and overshoot, while assessing cognitive load via the NASA-TLX survey and physiological sensors (blink rate, GSR, and thermal imaging).

The touchscreen interface outperformed the joystick across all key performance metrics. Participants using the touchscreen completed tasks 53.5% faster (median 2.50 vs. 5.38 minutes) and achieved higher path coverage (90.7% vs. 84.1%) with significantly less overshoot. Furthermore, the subjective cognitive load reported by participants dropped by 17.3% when switching from the joystick to the touchscreen. While the autonomous mode yielded the lowest cognitive load, the touchscreen provided a superior balance of human control and efficiency for tasks requiring active operator input.