Design and Modeling of a Smart Torque-Adjustable Rotary Electroadhesive Clutch for Application in Human-Robot Interaction

arXiv:2210.08664 · cs.RO, cs.SY, eess.SY · Submitted 2022-10-16 · 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: I'm Rosa.

Dev: And I'm Dev, here with Taro.

Rosa: We're looking at the paper: "Design and Modeling of a Smart Torque-Adjustable Rotary Electroadhesive Clutch for Application in Human-Robot Interaction".

Title and authors: Rosa: So we're looking at this paper today called "Design and Modeling of a Smart Torque-Adjustable Rotary Electroadhesive Clutch for Application in Human-Robot Interaction". It focuses on making a clutch that can adjust its torque using electroadhesive materials, which is pretty novel.

Dev: Yeah, it’s tackling the safety and workspace sharing issue between robots and humans by proposing this new type of clutch. The authors are looking at how to build something lightweight and low-power instead of the heavy magnetic or ER clutches we usually see.

Taro: I'm curious if this design actually works outside a lab setting, Rosa. Can we get real-world feedback on its durability when things get messy?

Rosa: Exactly, that’s what I want to know. The paper presents the design and modeling of this rotary electroadhesive clutch for use in human-robot interaction scenarios where safety is key. They're showing how you can have a controllable clutch without those big heavy ferromagnetic parts or complex fluids.

Dev: Right, because those existing clutches are often too power-hungry and too heavy for wearable or mobile applications twenty-one– twenty-three <ref:2210.08664#pg2>. This new approach uses electroadhesive force instead of magnetism, which requires very little energy to generate an electric field, which is a big plus for low-power systems.

Taro: Low power is good, but what about the actual performance? Does this clutch offer enough torque adjustment capability for the tasks we need to do in shared spaces?

Rosa: The paper investigates three different pairs of clutch plates to look at how smooth and quality of output torque are. They found that by using electroadhesive materials, you can get a much better torqueto-weight ratio compared to common adjustable clutches.

Dev: That's the main selling point for me, because achieving that ratio is critical for making these devices practical. The authors show that this clutch demonstrates a torquepower consumption ratio of six times more than a commercial magnetic particle clutch, which is quite significant.

Taro: Six times better sounds promising, but what about the control aspect? How does the system handle situations where things go wrong during operation?

Rosa: They addressed an issue with charge build-up and found that activating the clutch with an alternating current waveform helps solve that problem. They selected a specific configuration, number three, for further development of their experimental setup.

Dev: Configuration number three is interesting because they designed the dielectric to be placed against steel, which they found showed better performance than previous configurations with no obvious issue regarding torque fluctuations during activation.

Taro: So it handles the charge build-up issue with AC activation and has a specific configuration that seems stable when things are moving, but what about wear and tear over time?

Rosa: The paper points out that another shortcoming of this electroadhesive clutch is that it generates a sound noise with AC activation, so they need to investigate how operative conditions like temperature and humidity affect its extensive use.

Dev: That noise is something we have to consider for dynamic closed-loop control because those models are essential for understanding the system's behavior under different real-world conditions. The paper also mentions that these models are essential for dynamic closed-loop control in human-robot interaction applications.

Taro: So, while the hardware design seems solid with configuration three, the operational environment and noise are still things we need to model before it can be fully deployed.

Rosa: That’s right. To wrap up on this paper, the main implication is that this novel design offers a much higher torqueto-weight ratio and torque-to-power ratio than common adjustable clutches, which suggests it could be suitable for developing safe passive actuators for lightweight and low-power consumption robotic systems like exoskeletons and robotic walkers.

Dev: And they also noted that the results are important because they inform the feasibility of increasing the number of clutch discs in parallel without a significant increase in the total weight of the clutch, which is a really practical design consideration.

Taro: I'm just thinking about how this moves us toward systems where we can actually put these controllable mechanisms into daily life safely and efficiently.

Rosa: That’s right. So that’s what we have on this paper today: the "Design and Modeling of a Smart Torque-Adjustable Rotary Electroadhesive Clutch for Application in Human-Robot Interaction". It shows a path toward lightweight, low-power torque control using electroadhesive materials.

Dev: We'll keep an eye out for how the authors model the dynamics under different activation signals next time we talk about this paper.

The paper's summary: Rosa: So basically, this paper is about designing and modeling a clutch that you can adjust its torque just by changing the electrical voltage you put in it using electroadhesive materials.

Dev: Yeah, it’s looking at how to make something lightweight and low-power for situations where robots need to share space with people.

Rosa: They show that this approach uses an electric field between charged particles to create a force, which is then used to adjust the torque smoothly.

Dev: The main thing they highlight is that this design gives you a much better torqueto-weight ratio compared to those heavy magnetic particle clutches we usually see on the market.

Rosa: They found that in their testing with configuration number three, this clutch could achieve a torquepower consumption ratio six times better than a commercial magnetic particle clutch.

Dev: That’s significant because it means you can get more work out of the same amount of energy, which is key for low-power robotic systems like exoskeletons.

Rosa: They also tackled a problem they found with AC activation waveforms to prevent charge build-up, suggesting that using an alternating current signal helps keep things running better over time compared to just using direct current.

Dev: That’s a practical point because if the system degrades under direct current activation, you need a way to control that wear and tear, especially in dynamic applications where things aren't always steady.

Rosa: They also mentioned that they found configuration number three, where the dielectric material is placed against steel, worked better than their earlier designs with less fluctuation in the output torque.

Dev: So it seems like they’ve narrowed down a specific physical setup that works well for performance and stability when you're running real tests.

Rosa: But there are still some things they need to figure out, like how the temperature and humidity might affect how this clutch performs over a long period of use in the real world.

Dev: And those environmental factors lead directly into the next part of their work, where they need to build computational models that can predict what happens when you activate it with either direct current or alternating current.

Rosa: Those models are crucial for building dynamic closed-loop control systems, which is exactly what you need when a robot is interacting with a person in motion.

Dev: So the next big step here is creating those predictive dynamics models so that the AI can actually control this clutch safely and effectively during complex interactions.

The paper's improvements: Rosa: So, the authors aren't just stopping there on the initial design; they’re suggesting a few ways to really improve this clutch system for future work.

Dev: Yeah, they point out that using an AI system could be used to adjust the activating voltage dynamically, which could get you that torqueto-weight ratio six times better than those magnetic particle clutches we talked about earlier.

Rosa: That’s a big deal because it means the AI can fine-tune the torque in real time based on what's happening in your environment.

Dev: And they also look at how to handle torque degradation over time, and they suggest using an alternating current waveform activation signal instead of just direct current activation to keep performance stable after a long period of use.

Rosa: That’s a good safety feature because DC activation seems to cause significant torque degradation, so switching to AC helps manage that wear and tear.

Dev: They also propose building a new computational model specifically to estimate the dynamics of the clutch torque for both direct current and alternating current activations, which they say is essential for building dynamic closed-loop control in human-robot interaction.

Rosa: That model is what lets you know how much torque the system will produce next, which is exactly what you need if you’re trying to keep a robot safe while it’s moving near a person.

Dev: They also focus on the physical design itself, specifically configuration number three, where they put the dielectric material against steel because they found it showed better performance with no obvious torque fluctuations compared to other setups.

Rosa: So it sounds like the next set of improvements focuses on making the system smarter with AI control and building better predictive models for safety.

Dev: Right, and while those are great steps, they also flag that a real-world test is going to be needed to see how this all holds up when things get messy outside of a controlled lab setting.

Conclusion: Rosa: So to wrap up on this paper, we’ve looked at how they designed and modeled a smart torque-adjustable rotary electroadhesive clutch for human-robot interaction scenarios where safety is key.

Dev: The core finding is that this electroadhesive design offers a much higher torqueto-weight ratio than common adjustable clutches, and it can achieve a torquepower consumption ratio six times better than those magnetic particle clutches.

Rosa: It seems like the main implication here is that this technology could be really useful for developing safe passive actuators needed for lightweight robotic systems, like exoskeletons or walkers.

Dev: And they showed that by using an alternating current waveform, you can address the charge build-up issue and improve performance compared to just using direct current activation.

Taro: I’m still thinking about those noise issues they mentioned with AC activation; that sound is something we have to figure out if we want this on a robot in a busy area.

Rosa: Exactly, the paper flags that temperature and humidity conditions need more investigation for extensive use of this electroadhesive technology.

Dev: That leads directly into their future work, where they need to develop those computational models so you can have dynamic closed-loop control when things are moving fast in human interaction.

Taro: So the next thing we’ll be looking at is how robust those dynamic models are when the world misbehaves and you need graceful recovery.

Rosa: Right, and that brings us to our next paper on how AI can improve torque controllability by adjusting activating voltage to achieve those better ratios.

Navid Feizi, S. Farokh Atashzar, Mehrdad R. Kermani, Rajni V. Patel

Canadian Surgical Technologies and Advanced Robotics · Lawson Health Research Institute · School of Biomedical Engineering, Western University

cs.RO, cs.SY, eess.SY

Submitted: 2022-10-16

Updated: 2022-10-16

Comments: submitted to IEEE T-MECH, 11 pages, 14 figures,

DOI: 10.1109/TMECH.2023.3259926

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

Importance score: 76/100

The gist: The design and development of a novel, lightweight, and low-power torque-adjustable rotary clutch using electroadhesive materials is presented.

Key concepts

Electroadhesion
This is an electrostatic force that occurs between two charged particles, regardless of whether they are conductors or insulators. It is the fundamental physical principle used to create the clutch mechanism by controlling the shear stress at their interface.
Rotary EA Clutch
The clutch consists of a rotor disc and a stator disc that slide against each other. Pressure springs maintain an infinitesimal air gap, allowing electroadhesive forces to generate adjustable torque for continuous rotation.
Torqueto-Weight Ratio
This ratio measures how much torque the clutch can produce relative to its own weight. The proposed clutch achieves a significantly higher ratio compared to standard adjustable clutches, which is crucial for lightweight robotic applications.

Terminology

Summary

The design and development of a novel, lightweight, and low-power torque-adjustable rotary clutch using electroadhesive materials is presented. p. 0

Motivation for the Research

The increasing need for sharing workspace and interactive physical tasks between robots and humans has raised concerns regarding safety of such operations. Controllers that consider the safety of the robotic system along with the transparency of interaction such as passivity controller [2]–[4] and wave variable controller [5] are two of the known methods that provide safety at the system control level. p. 0

Electroadhesion Concepts

An electroadhesive force is an electrostatic force between two charged particles, whether they are conductors or insulators. The shear stress on the interface of the electrodes is governed by Columb’s law [36] as shown below. p. 1

Clutch Design and Actuation Scheme

The core of the proposed rotary EA clutch consists of a rotor disc (outer disc) and a stator disc (inner disc) sliding against each other. To address competing design requirements, pressure springs were integrated into the design to continuously push the stator disc against the rotor disc with an infinitesimal air gap between the discs. p. 2 The clutch was driven using a DC gear motor (Micro-Drives M4870U with 1:294 gear ratio) controlled by a Maxon EPOS2 under speed control mode with a constant speed. p. 3

Performance Validation and Modeling

The performance of the selected pair of clutch discs, configuration number 3, was evaluated through a series of tests using both AC and DC activations. The proposed rotary EA clutch offers a much higher torqueto-weight ratio and torque-to-power ratio than common adjustable clutches. p. 8 p. 8 The proposed clutch demonstrated a torquepower consumption ratio of six times more than a commercial magnetic particle clutch. p. 9 p. 9

Key Findings and Conclusions

The work addressed the charge build-up issue by activating the clutch with an AC waveform. Configuration number 3 was selected for further investigation and further development of the experimental setup. p. 4 The proposed rotary EA clutch offers a much higher torqueto-weight ratio and torque-to-power ratio than common adjustable clutches. The results are important as they inform the feasibility of increasing the number of clutch discs in parallel [30] without a significant increase in the total weight of the clutch. p. 8

Limitations and Future Work

Another shortcoming of the proposed EA clutch is that it generates a sound noise with AC activation. The effect of operative conditions such as temperature and humidity should be investigated for extensive use of this technology. p. 8 Such models are essential for dynamic closedloop control. p. 9 p. 9 The proposed clutch demonstrated a torquepower consumption ratio of six times more than a commercial magnetic particle clutch. Such models are essential for dynamic closedloop control.

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. Improved AI systems can integrate a novel rotary electroadhesive clutch to provide torque controllability by adjusting activating voltage, which allows for a torqueto-weight ratio six times better than commercial magnetic particle clutches.

  2. The system can address torque degradation with time and residual torque after deactivation by utilizing an alternating current waveform activation signal, leading to improved performance compared to DC activation which shows significant degradation under DC activation.

  3. A new computational model can be proposed to estimate the dynamics of the clutch torque for both direct and alternating current activations, which is essential for dynamic closed-loop control in human-robot interaction applications.

  4. The system can achieve a torqueto-power consumption ratio six times more than commercial magnetic particle clutches, making it suitable for developing safe passive actuators such as those required for lightweight and low-power consumption robotic systems like exoskeletons and robotic walkers.

  5. The clutch can be designed with a novel material configuration, such as configuration number 3 where the dielectric is placed against steel, which shows better performance with no obvious issue regarding torque fluctuations compared to previous configurations.

Related papers