MR-UBi: Mixed Reality-Based Underwater Robot Arm Teleoperation System with Reaction Torque Indicator via Bilateral Control

arXiv:2510.20407 · cs.RO · Submitted 2025-10-23 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Robotics Radio. Generated commentary on the latest robotics and control papers.

Rosa: We're looking at the paper: "MR-UBi: Mixed Reality-Based Underwater Robot Arm Teleoperation System with Reaction Torque Indicator via Bilateral Control".

Paper summary: Rosa: So, looking at this paper "MR-UBi: Mixed Reality-Based Underwater Robot Arm Teleoperation System with Reaction Torque Indicator via Bilateral Control," what we're seeing is a system that uses mixed reality to give operators real-time visual and haptic feedback on force application underwater.

Dev: The authors developed this MR-UBi by combining a low cost three degrees of freedom underwater robotic arm with a leader–follower bilateral control mechanism that gives haptic feedback, plus an RTI module for the visual cues.

Taro: What they really drove was that the reaction torque indicator, which shows torque through bar length and color hue, improved grasping accuracy significantly over their baseline system.

Rosa: They found that this meant operators spent more time in the optimal torque zone and experienced less error when handling objects with varying stiffness during tasks like lifting or picking things up.

Dev: Beyond the control performance, they confirmed usability metrics where the System Usability Scale score went up to a median of eighty-three point seven five, and NASA–TLX scores showed lower workload for the operators compared to the baseline.

Taro: It means that when you add this kind of intuitive feedback loop, you can actually make underwater manipulation tasks feel more stable and less mentally taxing for the person running it.

Rosa: So, in simple terms, MR-UBi is about using mixed reality and force feedback to make controlling underwater robot arms more intuitive and reliable for real work.

Conclusion: Rosa: So we're wrapping up on this MR-UBi paper, looking at the title and who wrote it, and what it actually means for us out here in the real world.

Dev: The authors are presenting this system called MR-UBi, which is basically a mixed reality setup for controlling underwater robot arms using a special reaction torque indicator through bilateral control.

Taro: What that boils down to is they've put visual and haptic feedback together to help people move these underwater arms better.

Rosa: It’s about taking the mess of force control underwater and making it more intuitive by showing the operator exactly how much torque they’re applying visually and physically.

Dev: The core idea is using that reaction torque indicator, which shows a bar length and a color to tell you if you're pushing too hard or not hard enough.

Taro: And the results showed that this actually works. They found it helps people stay in the right torque zone longer, which means less under-grasping and less over-grasping when handling things underwater.

Rosa: That’s a big deal because underwater tasks are inherently tricky with object stiffness, so if you can regulate that force better, you can handle much more complex stuff.

Dev: The usability scores they got were pretty high, indicating that this isn't just a neat trick in the lab; it’s something people actually want to use when they’re running these operations.

Taro: It suggests that integrating visual and haptic feedback through mixed reality could be a real way to make underwater manipulation safer and more reliable for actual deployment.

Rosa: Yeah, so this isn't just theoretical control anymore; it’s showing how we can build tools that are genuinely user-friendly for tough environments.

Dev: It opens up the door for future work looking at how long these systems can run reliably in those real underwater conditions without breaking down.

Kohei Nishi, Masato Kobayashi, Yuki Uranishi

University of Osaka · Kobe University

cs.RO

Submitted: 2025-10-23

Updated: 2025-10-23

Journal ref: IEEE Access, vol. 14, pp. 40137-40146, 2026

DOI: 10.1109/ACCESS.2026.3672494

Project page: https://mertcookimg.github.io/mr-ubi

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 80/100

The gist: MR-UBi presents a mixed reality-based underwater robot arm teleoperation system with a reaction torque indicator via bilateral control <ref:2510.20407#pg2> The reaction torque indicator (RTI)…

Key concepts

MR-UBi
A mixed reality system designed for teleoperating underwater robot arms. It combines a physical robotic arm with a Mixed Reality Head-Mounted Display (MR-HMD) that provides visual and haptic feedback, allowing remote control through this immersive interface.
Reaction Torque Indicator (RTI)
A visual cue overlaid in the MR headset that represents the robot's reaction torque. It uses both bar length to show how strong the force is and color to indicate if the applied torque is in a low, optimal, or high range.
Bilateral Control
A control mechanism used for leader-follower synchronization between two systems. In this context, it enables haptic feedback, allowing an operator's movements to influence the robot arm and receive tactile sensations back through the control loop.

Terminology

Summary

MR-UBi presents a mixed reality-based underwater robot arm teleoperation system with a reaction torque indicator via bilateral control <ref:2510. p. 020407#pg2> The reaction torque indicator (RTI) overlays a color and length-coded torque bar in the MR-HMD, enabling seamless integration of visual and haptic feedback during underwater robot arm teleoperation. p. 0 User studies with sixteen participants compared MR-UBi against a bilateral-control baseline MR-UBi significantly improved grasping-torque control accuracy, increasing the time within the optimal torque range and reducing both low and high grasping torque range during lift and pick-and-place tasks with objects of different stiffness Subjective evaluations further showed higher usability (SUS) and lower workload (NASA–TLX) Overall, the results confirm that MR-UBi enables more stable, accurate, and user-friendly underwater robot-arm teleoperation through the integration of visual and haptic feedback. p. 0

System Architecture

MR-UBi comprises three modules: (1) a low-cost 3-DoF underwater robotic arm, (2) leader–follower bilateral control mechanism that provides haptic feedback, and (3) RTI module that enables haptic-based visual feedback via MRHMD passthrough. The system integrates 3-DoF underwater robot arms, four-channel bilateral control, and MR-HMD passthrough with RTI. p. 1 The leader–follower synchronization is achieved through a four-channel bilateral control algorithm. p. 2

Reaction Torque Indicator Design

The RTI represents reaction torque using a dual representation scheme that combines two types of visual cues: (1) the bar length, which shows the torque magnitude quantitatively, and (2) the color hue, which indicates whether the torque is within an appropriate range. The bar filling ratio L(τ) is calculated by normalizing the reaction torque τ within the range between its minimum Tmin and maximum Tmax values. p. 3 The torque range is divided into three range: Low (τ ≤ Tlow), optimal (Tlow ≤ τ ≤ Thigh), and High (Thigh ≤ τ), each associated with a representative color: Clow, Copt, and Chigh, respectively. p. 3 A transition margin Mtot is introduced to achieve smooth color blending between neighboring range. p. 3

Experimental Validation

The experimental results confirmed both objectives by showing that MRUBi improved grasping-torque control accuracy by increasing the time within the optimal torque range and reducing torque error. The RTI curbs under/over-grasping and helps operators rapidly identify and maintain the appropriate torque range. p. 6 The reduction of excessive torque with the compliant object is especially notable, highlighting the RTI’s effectiveness in suppressing over-force events. p. 6

User Experience Metrics

The System Usability Scale (SUS) score under MR-UBi increased to a median of 83.75, a significant improvement over Baseline. p. 6

Conclusion

This paper presented MR-UBi, a novel mixed reality-based underwater robot arm teleoperation system with reaction torque indicator via bilateral control. By encoding reaction torque through the RTI, the MR-UBi enables intuitive force regulation even in underwater robot arm teleoperation. p. 7 User studies with sixteen participants performing two manipulation tasks and objects demonstrated that MR-UBi significantly increased time within the optimal torque range and reduced both under- and over-grasping, while SUS and NASA–TLX results confirmed improved usability, confidence, and reduced workload. p. 7 These findings verify that visual–haptic integration via MR-UBi facilitates stable and intuitive underwater robot arm teleoperation. p. 7 The experimental results validate the overall effectiveness of MR-UBi in real-world underwater tasks. p. 7

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improve underwater robot arm teleoperation stability by integrating a reaction torque indicator (RTI) that encodes torque magnitude via bar length and state via a hybrid continuous discrete hue scheme. This allows operators to perceive gradual torque changes smoothly, while still clearly distinguishing optimal range.

  2. Enhance grasping-torque control accuracy by increasing the time within the optimal torque range and reducing both low and high grasping torque during lift and pick-and-place tasks with objects of different stiffness, as shown in Table III results.

  3. Improve user experience by significantly boosting perceived usability, confirmed by a median System Usability Scale score increase from 66.25 to 83.75 with MR-UBi, indicating that the RTI boosts confidence while reducing hassle and stress.

  4. Reduce operator workload during underwater manipulation tasks, as evidenced by a lower NASA Task Load Index score under MR-UBi (41.17) compared to the baseline (49.50), showing lower physical effort and effort-related burden.

Sources

Related papers