Piloting a Remotely Operated Vehicle (ROV) through a hazardous underwater environment means fighting currents you often can’t see coming, with no realistic prospect of recovering the vehicle if something goes wrong. A new study, accepted for IEEE Telepresence 2026, tests whether a simple haptic touchpad can help an operator feel a threat they can’t visually track — and finds a more complicated answer than “yes, it helps.”
The setup
Researchers ran an in-water study exposing an ROV to controlled cross-currents, with environmental information relayed to the pilot through a soft, vibrating haptic touchpad. The comparison is specifically between navigating an unseen current with no haptic information at all versus navigating the same current with haptic cues layered in — isolating the haptic channel’s effect rather than comparing haptic feedback against some other sensing modality.
Faster, more confident — and more frustrated
The results split in an interesting direction. Encountering an unseen cross-current significantly increased operators’ physical and temporal demand regardless of condition, which is the expected baseline cost of an unpredictable hazard. Adding haptic feedback mitigated those demands and improved self-assessed performance — operators felt like they were doing a better job navigating the hazard with the haptic cues present than without them.
But the haptic condition also produced the highest levels of mental demand and operator frustration measured across the study, even though it reduced physical and temporal demand. That’s a real tension rather than a clean win: the touchpad gave operators more information to act on, which apparently took real cognitive work to interpret even as it paid off. Despite that added mental load, operators piloting with haptic cues drove the ROV significantly faster through the cross-currents than those in the current-only condition — read by the authors as evidence of higher operator confidence, since driving faster through an unseen hazard is a confidence-dependent choice, not a free one.
The honest reading
The authors’ own framing is careful rather than triumphant: haptic feedback may initially demand more cognitive processing to interpret an unseen hazard, but it still enhances operator awareness and promotes more confident navigation in exactly the conditions — hazardous, currency-prone, visually incomplete — where that confidence actually matters. For interface designers working on haptic feedback anywhere outside this specific underwater case, the finding worth carrying over is the tradeoff itself: a new feedback channel can improve measured performance while genuinely costing the operator something in frustration, and a system that only tracks task outcomes will miss that cost entirely.
The paper is accepted for IEEE Telepresence 2026, giving it a confirmed peer-reviewed venue rather than standing as an unreviewed preprint.