Everything you need to know about bionic arms and hands, with descriptions of all the current devices, technologies, and the latest research. Continuously updated, this page is intended as your one-stop repository for all the latest information on upper-limb bionics.
What’s On This Page?
- The Need for Bionic Arms & Hands
- How Bionic Hands Work
- Below-the-Elbow Bionic Arms/Hands
- Partial Hand or Finger Options
- Above-the-Elbow Bionic Arms
- Latest Technology Articles for Bionic Arms/Hands
- Latest Research for Bionic Arms/Hands
- Bionic Feats by Bionic Arms/Hands
- Real Stories About Bionic Arms/Hands
- Related Information
The Need for Bionic Arms & Hands
According to recent WHO data, roughly 3 million people live with upper limb amputations.
These amputations can occur at several points, with the vast majority involving one or more fingers. Approximately 14 % of upper limb amputations occur below the elbow or at the wrist. Another 21 % occur at one of the 3 points above the elbow.
Most upper-limb amputations are due to physical trauma (70 %), with disease, infection, and congenital disorders causing the rest.
Until the early 2000s, the only treatment options for this type of loss were:
- passive prostheses, ranging from cosmetic arms/hands with little or no functionality to rigid tools, such as a hook;
- body-powered devices, which use harnesses or cables for added movement/capabilities;
Unfortunately, these traditional solutions are often rejected by patients due to discomfort, a lack of functionality, and the stigma of limb loss.
Human bionics offer solutions to all these problems.
How Bionic Hands Work
The best way to understand how bionic hands function is to start with the fingers and work backward.
The core mechanical function of a human finger is to open and close. This is replicated in a bionic finger using a design like this:
A battery-powered motor drives a gearing system to move the main MCP joint. This, in turn, moves the second and third joints via a bar linkage system. Note, there are many variations of this, some using pulley systems to simulate tendons instead of bar linkage systems.
The key feature in this type of design is that there is typically only one independent joint: the MCP joint (main knuckle). The other two joints automatically follow that joint’s lead, producing the following fixed pattern of motion:
By comparison, natural hands can move the middle or PIP joint independently, and the DIP joint quasi-independently. Additionally, natural fingers can spread out and twist and turn in a myriad of combinations.
The movement of a natural thumb is even more challenging to duplicate. In addition to opening and closing, it can rotate down and slightly outward to touch the entire underside of any finger and even parts of the palm. This allows natural hands to grasp objects of any shape.
Here is a diagram of a bionic thumb from the early work of one of the more recent entries in the bionic hand market, Psyonic.
Despite its mechanical complexity, this type of thumb still has many limitations. Yes, it can position itself to become an opposable force for many grips, but it lacks the exceptional dexterity of a natural thumb.
And that exemplifies one of the main points of this section: bionic hands are impressive feats of engineering, but they are much simpler than natural hands. In one respect, this hardly matters because they can still perform a wide variety of tasks, as demonstrated in this video:
But simplification does come at a price…
Automated Control Issues
One of the grips shown in the preceding video is the pinch grip, where the thumb and the forefinger come together to pick up a small object.
If you attempt this grip with a bionic hand, and the two digits are misaligned, it won’t work. Also, sufficient force must be exerted to pick up the object, but not so much as to crush it if it is fragile, like an egg. Finally, using this grip to carry an object of significant weight requires continuous pressure, but it is inefficient to ask the motor to apply this pressure, so some kind of automated locking system must be used instead.
These and/or other variables apply to each grip. Accommodating these variables in a simplified electromechanical hand involves some compromise compared to our natural hands. And as you might expect, some design teams deal with these challenges more effectively than others.
This is the crux of bionic hand design, or at least the electromechanical aspects of it. But even if a bionic hand is capable of performing a task, there is still the question of how you control it.
Myoelectric Control Systems
When you move a natural limb, the brain sends nerve signals to the muscles, which in turn move the limb.
When a limb is amputated, the brain still sends these signals even though some of the muscles are no longer there to receive them.
One way to control a bionic hand is to directly interpret the nerve signals, but this requires surgery to implant electrodes. This is not only invasive and expensive; it also risks complications such as infection.
A non-invasive alternative is to focus instead on the movements of the remaining muscles. Muscle movements generate their own electrical signals, which are stronger and more easily detected on the skin’s surface than nerve signals.
Most bionic hands rely on these surface signals to enable user control. This is what is referred to as a “myoelectric control system”. In the simplest model, two electrodes are placed against the skin on the residual limb. These are calibrated to detect 1) an attempt to open or extend the missing hand, and 2) a closing/curling motion.
These signals are then mapped to the corresponding open and close functions of the bionic hand via a control system.
Exactly what is being opened or closed depends on the hand’s current grip selection. For a pinch grip, it is just the forefinger and thumb. For a hook grip (used to carry a shopping bag or briefcase), all four fingers must open and close in unison.
To take advantage of multiple grips, users need a way to switch between them. One common method is to allow users to cycle between grips by repeatedly extending the bionic hand (i.e. repeated open motions). Another method is to map specific trigger movements (e.g. double open impulse, co-contraction, etc.) to specific grips. Buttons, Bluetooth mobile apps, and grip chips (chips that trigger specific grips when you pass your hand within a certain distance of them) are also used.
Some bionic hands offer even more sophisticated control systems. Using as many as 8 sensors in combination with artificial intelligence, they map patterns of muscle movements to specific grips. One company, Brainco, even allows you to control individual fingers without pre-programmed grips.
But caution is in order when assessing these control systems. Some bionic arm recipients have less residual muscle than others, meaning they must learn to trigger open and close movements using muscles that are not naturally prominent in these actions. Some people are good at doing this and some are not.
Put another way, the ability to generate the muscle signals needed for myoelectric control can vary significantly from person to person.
Even if someone is adept at generating the proper signals, the myoelectric sensors may not always detect them. Some sensor systems may work okay in cool, dry conditions, but malfunction if the arm’s socket/shell becomes hot and sweaty. Others may work well when the arm is horizontal, but not if it is bent at a significant angle.
For more information on these issues, see finding the right myoelectric control system. For more information on control systems in general, including myoelectric alternatives, see mind-controlled bionic limbs.
Below-the-Elbow Bionic Arms/Hands
The following is a list of the models for below-the-elbow bionic arms/hands either already on the market or soon to be (presented alphabetically). Each links to its own page for more specific details:
- Aether Biomedical Zeus Hand
- BrainCo Dexus Prosthetic Hand
- Open Bionics Hero Arm
- Ossur i-Limb
- Ottobock Bebionic Hand
- Ottobock Michangelo Prosthetic Hand
- Psyonic Ability Hand
- Taska Prosthetics
- Unlimited Tomorrow TrueLimb
- Vincent Evolution 3
Of the bionic hands currently on the market, the most affordable are the TrueLimb from Unlimited Tomorrow, the Hero Arm from Open Bionics, and the Zeus Hand from Aether Biomedical. In their simplest configuration (least complicated residual limb), the final price for these three devices can range from $8,000 US to $15,000, including prosthetist fees, where applicable.
By comparison, the most expensive hands can cost more than $60,000 USD.
For more information, see our Bionic Hand Price List.
Partial Hand or Finger Options
As pointed out in The Need for Bionic Arms & Hands, nearly 2/3 of upper limb amputations involve finger or partial hand loss. This equates to millions of people around the world who are missing fingers, parts of fingers, or a part of their palm.
Both Ossur and Vincent Systems have created excellent bionic partial hand prostheses, but these are not well-suited for heavy loads or challenging environments. They’re also quite expensive. To address these deficiencies, our list of articles on partial-hand devices includes the best non-bionic (i.e. strictly mechanical) options:
In keeping with our goal of helping you stay informed, we have also created a Current Partial Hand Prosthesis Options summary page. We will constantly update this page so that you can use it as a convenient place to stay informed about all the latest partial hand devices.
Above-the-Elbow Bionic Arms
In general, the further up the arm an amputation occurs, the more complex the requirements for a bionic prosthesis. This is due in part to the need for additional moving joints. But weight and power consumption also increase, as does the requirement for a stronger attachment to the residual limb or body.
Solving these problems required a bigger budget and more technical collaboration.
Enter the Defense Advanced Research Projects Agency (DARPA). In 2006, the agency launched its Revolutionizing Prosthetics program. The goal of this program was to develop a bionic arm to dramatically improve the quality of life for upper limb amputees. They wanted an arm capable of mimicking the natural arm and hand movements for any level of amputation.
Two advanced bionic arms resulted from this program.
The LUKE Arm
The LUKE Arm was developed for DARPA by DEKA Research and Development Corporation.
In 2014, the FDA approved the arm for commercial use:
As you can see in the video, the arm clearly fulfills its original goal. It can indeed mimic most natural arm and hand movements. In the preceding video, the wearer uses foot controls like joysticks to manipulate the arm.
However, myoelectric controls can be used here, too.
Modular Prosthetic Limb (MPL)
The Modular Prosthetic Limb (MPL) was developed for DARPA by Johns Hopkins University (JHU). This was a more complex hand/arm system designed to “test direct neural control of a prosthesis”.
It has since made its way out of the lab and into home trials.
As the video demonstrates, the ability to manipulate this arm via thought is quite sophisticated. But it still isn’t the same as a natural arm/hand.
To make the MPL and the LUKE Arm behave more naturally requires a) a more sophisticated neural interface, and b) the incorporation of touch. To read more about these efforts, see Mind-Controlled Bionics and Bionic Touch.
Ottobock has offered the DynamicArm for sale since late 2009. Although its technical origins are not entirely clear, it appears to be an above-the-elbow version of the myoelectric arms that have become so popular for below-the-elbow amputees. Put another way, it is a highly effective bionic arm for the above-the-elbow amputees, but it does not have the sophistication of either the LUKE arm or the MPL.
Here it is in action:
The big problem with above-the-elbow bionic arms is cost. The LUKE Arm costs as much as $100,000 USD. The cost for the MPL, should it get to market, will likely be similar. Even the DynamicArm, commercially available for over a decade, is still around $60,000 USD.
Unfortunately, this price tag exceeds the budget of most amputees. However, we should not be discouraged by this. It takes a while for new technologies to trickle down into the broader market. And then it takes a while longer (i.e. the effect of market forces) to significantly reduce costs. We just have to do what we can to accelerate these processes.
In the interest of accuracy, both the LUKE Arm and the MPL are not strictly above-the-elbow prostheses. Their modular design allows them to be used for all levels of upper limb amputation.
Latest Technology Articles for Bionic Arms / Hands
Latest Research Articles for Bionic Arms / Hands
Myoelectric control systems form the interface between bionic arms / hands and their human users. Finding the right control system is one of the most important factors in user satisfaction.
Bionic Feats for Bionic Arms / Hands
Real Stories for Bionic Arms / Hands
The innovators featured in this post are all accomplished scientists, engineers, and/or inventors. They have all made significant contributions to the advancement of bionic limbs. But what makes them truly special is their passion to improve the lives of those…
Bionic arms can attach directly to the humerus (upper arm bone) or radius and ulna (forearm bones) through Osseointegration.
This improves range of motion, strength, stability, and also adds a rudimentary sense of touch (through vibration). For more information, see Osseointegration for Bionic Limbs.