Questions You Should Know about exoskeleton joint actuator

Hi everyone,

ARCSEC DRIVE Product Page

I would like to become more active in the field of exoskeletons - probably building an exoskeleton. The questions are somehow always the same with every such project: Why? How come? What for? Pourquoi?
You could help me, building 'the right' exoskeleton. I just would like to find out, what makes sense most. At least I would like to avoid, just building an exoskeleton only for myself. Probably it could be useful for some other people as well.

That's why I've created a short survey (<7min) to shed some light on it. I would be very happy if you could take part in it and give me feedback:
https://yujp90be53w.typeform.com/to/mOvytIhT

I am very grateful for your feedback!
Of course, if you are interested I could present the results here.

Best regards
Enrico

Hi all,

as promised, here are the results of the survey. First of all: thanks to everyone, who participated in it. It was very helpful for me.
These are the direct answers to the questions. So no interpretation or correlation included here.
Let's start with the Demographics:

• in total, there were 25 participants in the survey
• the average age was 42
• 84% declared to live in europe, the others did not mention their current location

How likely is it that the participant or a friend/relative of the participant would use exoskeletons?

Which limb/joint would be of most interest to be supported?

• 8% of participants would like to have a full body exoskeleton
• Besides, 12% would like to have a full set of legs
• 8% are interested in full hands (wrist & fingers)
• 72% are interested in (multiple) single joint solutions
• 0% wanted full arm exoskeletons

• Exoskeletons for the knee most favorable (64%)
• Hip and Foot ankle close by
• Hand and back equal
• Arm (Fingers, Elbow, Shoulder) on wish list of '20% of participants

For which activities would exoskeleton be used?

• Just for fun/Cosplay/Accessory not of interest
• As Sports device not very interesting
• Most probably used, when advantage for work & health is expected

• Motor assistance most important for exoskeletons ' full replacement of muscle activity demanded
• Muscle activity sensing important
• Design important, but no need for 'cool LEDs'
• Controllability almost equal: RC, App, Interface to other processors or microcontrollers

Tradeoffs: Speed vs. Strength, Weight vs. Power, Energy vs. Weight

• Mostly rated with '3'
  The questions seem too technical to me. Most people don't know how much force/how much speed/how lightweight/'
  It might depend on specific use case & kind of exoskeleton as well as on specific numbers (e.g. 15mins/2hrs/8hrs of operation vs. specific weight)

Willingness to pay for quality and durability

• People prefer adequate, well built devices for sports, as a working aid or as a limb support, or
• to some amount cheap exoskeletons just for fun
• 23% would not spend money for exoskeletons

Additional Comments
These were the comments, that the participants left

I hope, that it is as informative for you as it was for me. If you have questions regarding these, just let me know.

Best regards,
Enrico

Hi again,

I used the raw data to also correlate some of the answers.
What I also found out: with the 'hand' exoskeleton, I was not very precise. What I actually meant is a wrist exoskeleton. That's why the fingers and thumb are listed as a seperate exoskeleton. I hope, that all participants saw it the same way - but I cannot guarantee for that.

Lets start:
Age vs. Likeliness to wear exoskeletons
How is the likeliness to wear exoskeletons depending on the age of the participant?

• no participants were younger than 20, so I skipped this line.
• the bottom diagram is the sum of both upper diagrams (so the total sum-up of all entries is 200% in that matrix)
• The higher the age, the more likely it is, that the person is willing to wear exoskeletons (or knows s.o. who might do so)
• older than 30yrs ' more willing to wear exoskeletons

Limb/Joint of most interest vs. its application
This matrix correlates the interest in a certain joint with the purpose that the participant prefers. E.g. if someone is only interested in 'back exoskeletons' and prefers to wear it 'Just for fun' (high rating) it would increase the score in that field in the table.

• body weaknesses: legs, especially knees seem to be the most favorable joints
• Working aid: all joints equal
• Shoulder, elbow, back not wanted for fun/Cosplay/Accessory
• Ankle, knee & shoulder might also be used just for fun or sports
• Fingers & hand interesting for fun and cosplay as well

Limb/Joint of most interest vs. required features
Question: can a certain feature be assigned to a certain type of exoskeleton or joint? E.g. do hand exoskeletons need LEDs and knee exoskeletons preferable strong motors? This diagram shows some trends about this:

• Motor assistance for all joints important, but if highest interest for the legs and elbows
• Sensing of muscle activity most interesting for legs and elbow
• Design for leg exoskeleton seem important (but cool LEDs are not necessary)

Tradeoffs in relation to the joints
Do some trends regarding high strength/high speed, lightweight/high power or large/small battery correlate more or less with some joints?

So that's it actually.
As I said: hopefully it is helpful for you and don't hesitate to discuss about it or ask me some further questions.
For me it was interesting to go through the process, because it is the first survey that I ever set up.

Best regards,
Enrico

I didn't see any indication that you have formal training in orthotics. If not, keep in mind that you are attaching your device to a human being, and it will likely be powerful enough to injure the person if something goes wrong. Motor controllers can fail in a way that the motor turns on unexpectedly, and of course, a programming error will likely resut in the same thing at some point in its development. So, you have to be pretty good at both biomedical engineering, and in designing ultra-reliable robotics. You've seen a need, and you are now hunting for a way to address that need, but you have chosen a very difficult project.

If you want to proceed, I suggest you abandon any idea of strapping anything onto anyone for now. Instead, focus on things that people sit on, or perhaps grab. E.g. there are chairs that will lift up the seat cushion to help a person get out of the chair. Of course, that is a problem that is already solved. But, perhaps you can find another.

I will mention a particular need I see, that is not yet met in the market. Personally, I have a muscle disorder and am easily fatigued. This is only a problem when I go to shopping malls, or to a county fair. I am not an invalid, so I don't need medical grade equipment to get around. If an scooter occasionally gets stuck when it is used on the grass, I can deal with that. There are some relatively low cost mobility scooters out there, which are both very light (under 40 pounds), and low cost (around $400-$700). But, those cannot go over grass because the single drive wheel just spins in the grass. [ VEVOR Portable 3-Wheel Mobility Scooter for Seniors 12 Mile Range Max 330LBS | eBay or Robot or human? ]. There are other ones out there, that are both light-weight and can be used in grass, but those cost dramatically more. Engineering is not just about calculating forces, but also economics.

The ones with a single front power wheel, cannot go up hill, because all your weight transfers to the unpowered rear wheels. The one with rear-wheel power, really needs both wheels to be powered. Also, you really need larger wheels, such as ones made for hoverboards, that can go over grass. In fact, hoverboards have turned powerful BLDC motors into commodity items.

My bottom line recommendation is to start with something simpler, to help build your robotics skills. And choose something that doesn't present a danger to its user until you gain the knowledge and experience to properly design a biomedical device.

Contact us to discuss your requirements of exoskeleton joint actuator. Our experienced sales team can help you identify the options that best suit your needs.

@fb1: I hope you have seen my pm?
@cadcoke5: Thanks for your suggestions. I can imagine, that it is very frustrating, to have only not-so-well-engineered devices available. Of course it would be a nice topic to redesign/optimize such a scooter.

I would prefer not to focus on low hanging fruits. For me arranging some wheels with motors and make them turn on command seem like a moderate challenge. But I guess, there are many people out there, who are quite capable of achieving such a functionality. Hereby, I would encourage everyone who is reading this to think of such a solution. Probably it is not only @cadcoke5 who would benefit from it.
Personnally, I would prefer something more challenging. I am not a biomedical engineer, but I think, engineering in this direction, learning about regulations and following them, might end up in a safe exoskeleton, that could one day be worn be someone. As you said, probably it is about engineering, and engineering a safe device that would not harm someone. Probably the biggest challenge in it lies in a good risk as well as failure mode and effects analysis - and of course the reduction of all the potential risks and failure modes. So I just try it this way. Of course the device would not be attached to any body (also not mine), if the device is not working properly.

But I thank you for your suggestions and look forward to many others.

Best regards
Enrico

Since your goal is to create a relatively novel approach to an exoskeleton, consider trying to design it so that any failure in controlling the motors are just not going to cause harm due to the nature of the design.

I gave the seat-lift idea as an example. Such devices use worm-drive actuators, so if a motor turns off, the user is not suddenly dropped down. And the motor system is designed so that its maximum speed cannot throw the user off the chair. Nor is the device even strapped to the person. Though, theoretically, the designer of such a system might have designed a lift that was strapped to the user's legs instead being inside the chair. There are also crane-type of lifts. Which help get people out of bed, as well as out of a chair. But, they are not as easy to use, and generally a person can't use it by themselves. There can be multiple ways to solve a problem, with some being safer or easier to use.

For your exoskeleton, I am not clear about your end-use goals. If it can be more narrowly defined, perhaps like the chair lift, it can lead to a more limited usage, but an easier and safer one. (i.e. the in-chair mechanism, vs. the crane type lift) For your exoskeleton, if the goal is to help users lift boxes and put them onto shelves, then the device may not need to strap to the user, It could be a somewhat independent robot, designed so that user grabs some handles on it, and directs the robot in its actions by moving those handles.

If we have an exoskeleton with a frame we have a static structure, but if we want to be able to walk we need actuators. An actuator converts energy into movement or force. In an exoskeleton, these actuators are called the joints. The main goal of the joints is to move the bones of the exoskeleton relatively to each other.

First of all, there are a lot of different types of joints that can be categorized in many different ways. As you can imagine, tiny drones or small robots need different actuators than large cranes in large cargo ports or in the automotive industry. Exoskeleton actuators should be able to deliver enough rotational force and speed but should also be small and lightweight. And even within these requirements, there are still quite some options. Not all possible actuators are explained in this reading, but the most common ones will be.

Passive and active joints

In exoskeletons, you can use passive and active joints. Active joints are actuated where passive joints are not. Passive joints are connection points between bones. With an axis, the bones can turn around, with optional non-actuated parts such as springs attached to them. You can't control the precise position of the joint within the outer limits. Active joints are more common in exoskeletons since you often want control over the position of limbs and therefore the joints. If a passive joint is used, it is often the ankle joint. You don't need to power this joint to be able to walk, although you lose some control over the exoskeleton position if you don't. Joints can move in different ways. In linear joints, two parts glide past one another without changing the angle between them and without rotation. The mobile part retracts and extends from its housing.

Linear and rotary joints

Joints can move in different ways. In linear joints, two parts glide past one another without changing the angle between them and without rotation. The mobile part retracts and extends from its housing.If a linear joint is placed around a passive rotating joint, it turns the passive joint into an active joint. By shortening and lengthening the linear joint, the bones start rotating around their connection point.

Figure: Motion of a linear joint

If a linear joint is placed around a passive rotating joint, it turns the passive joint into an active joint. By shortening and lengthening the linear joint, the bones start rotating around their connection point.

Figure: Motion of an actuated linear joint and passive rotational joint

As the name already suggests, rotary joints make the bones on both sides rotate respectively to each other.

Figure: Motion of a rotary joint

Although these rotary joints might look more similar to the joints in our body, our body actually uses the same mechanism as the combination of a linear joint and a passive rotary joint. The joints in our body are simply connections between bones that allow movement in certain directions. By contracting and releasing the muscles that are connected to the bones on either side you are able to move your limbs! In exoskeletons, rotary joints are used more often. They can deliver enough rotational force and speed, but with a larger range of motion and a more compact design compared to linear joints.

In the MARCH IV exoskeleton, both rotational and linear joints are used. As you can see in the picture below, the knees and hips are equipped with rotary joints. The hips also have linear joints for a sideways motion, and so do the ankles.

            
Figure: Render of the March IV exoskeleton

Electric, hydraulic and pneumatic actuators

Actuators can be powered by electricity, but also by pressurized air or fluids. All these types have their advantages and disadvantages. At this moment, most exoskeletons use electric actuators.

Pneumatic actuators are powered by pressurized air. By moving compressed air to different compartments within a joint, the position of the joint can be determined. Pneumatic actuators are quite safe since they do not contain any hazardous materials, although high pressure itself can always lead to unsafe situations. On the other hand, they deliver little force to power the exoskeleton (it is possible though) and they have quite a low efficiency due to air leakage. If you want to accurately control the actuator, you need regulators and valves, which complicate the design and increase the weight, dimensions, and costs.

Figure: Pneumatic or hydraulic linear actuator

Hydraulic systems operate quite similar to pneumatic systems, but with the use of fluids instead of air. They can generate very high forces. Although there is a loss of efficiency over time due to leakage of fluid. Leakage of fluids can also cause damage to the surrounding systems. It requires reservoirs, pumps, valves, and more to be able to have precise control over the actuator and therefore greatly increase the complexity and weight of the exoskeleton.

Electric actuators offer very precise control and are quick and easy to use. They can generate quite some power and are more quiet compared to both hydraulic and pneumatic systems. They are relatively compact and lightweight, which makes them suitable for exoskeleton use.

Electric motors and transmission

When using electrical actuators, there is a difference between the motor and the transmission. The motor actually uses electricity and converts it into movement, or kinetic energy. Transmission converts it into the movements that you actually need. On the next page, there will be a video with some explanation on different types of transmission and how you can do some calculations on transmission. In exoskeletons, it is very common to use a drone motor or another small motor that has a high rotational frequency, combined with a transmission that slows down the speed and increases the rotational force.

You can also combine hydraulic, pneumatic, and electrical systems. You can, for example, use an electric motor and combine it with a hydraulic transmission system. The benefits of combinations like this are that you get to combine favorable properties of both systems and reduce the disadvantages.

Overall, there are many different types of actuators for a great range of applications. By being precise in establishing your list of requirements you should be able to pick the right joint for your exoskeleton.

Are you interested in learning more about Robot Joint Actuator? Contact us today to secure an expert consultation!