Listen to this article
Pushed forward by the emergence of more powerful and more compact robotic technologies, biomechatronic-based prosthetics have made significant gains over the past few years. The market has also been assisted by the U.S. Defense Department, which has invested millions of dollars into multiple research projects designed to aid wounded warriors.
Unlike non-robotic artificial limbs, typically made out of plaster and steel, biomechatronic prosthetics take advantage pf stronger, lighter, and more flexible composite materials, which make them more natural to wear. For example, the state-of-the-art MIT BiOM T2 ankle-foot prosthesis weighs just 5.1 lb., which is equal to or less than the weight of its anatomical equivalent.
Biomechatronic prostheses use a combination of sophisticated technologies. Biosensors, which receive signals from a user’s nervous or muscle system, detect a person’s intention to move in a certain way. Mechanical sensors pick up information about the prosthesis’ current position.
Both types of sensors connect to a microprocessor-powered controller that acts as an interface between biological and electronic systems while also monitoring and adjusting the prosthesis’ movements. Actuators do the grunt work, serving as muscles to move an artificial limb.
Biomechatronic researchers pay close attention to developments in a variety of fields, including electronics, computers, and mechanical engineering.
“We watch battery technology, for example, because everything we do has to be self-powered and worn on the person,” said Richard Weir, director of the Biomechatronics Development Laboratory at the University of Colorado, Denver. “We are always interested in lightweight, higher-density power packs; better electronics; and lightweight, small, and more powerful motors.”
Researchers at Baltimore’s John Hopkins University have worked for nearly a decade to develop a prosthetic arm and hand that they hope will eventually restore everyday functionality to people who have lost a limb to disease or trauma.
The project began in 2005 when the Defense Advanced Research Projects Agency (DARPA) put out a call for technology to restore natural limb function to individuals who had suffered amputations in the line of duty. DARPA was searching for prosthetics that would look, feel, weigh, perform, and seamlessly integrate with a human user as if it were a natural limb.
The Modular Prosthetic Limb, developed by a research team led Michael McLoughlin, chief engineer of research and exploratory development at Johns Hopkins’ Applied Physics Laboratory, features an anthropomorphic form factor and appearance as well as human-like strength and dexterity.
Other attributes include high-resolution tactile and position-sensing capabilities and a neural interface for intuitive and natural closed-loop control.
The device is designed to be suitable for use by people who have lost an entire arm or leg.
McLoughlin noted that DARPA’s involvement was essential for spurring technology development in a largely underserved market sector.
“A replacement for the human arm is much more complex than for a leg, yet you don’t have as many people who need it, so you don’t have the commercial pull to stimulate the technology’s development,” he explained.
The prosthesis incorporates more than 100 sensors. McLoughlin noted that many candidate technologies were prototyped and evaluated using a custom-designed test bed and various standardized processes.
Design constraints included integration, reliability, and manufacturability. At the arm’s individual joints, sensors are used to measure angle, velocity and torque.
Additional sensors, located at the fingertips, measure force, vibration, fine-point contact, and temperature/heat flux.
To reduce design complexity, the team needed to design motor controllers that were usable at multiple joint locations. A large motor controller (LMC) was developed to provide a circuit design that could be leveraged for use at the four joints of the upper arm and the three joints at the wrist.
The LMC offers brushless direct current (BLDC) motor commutation, sensor signal sampling, and communication with the central limb controller (LC) via a controller-area network (CAN) bus.
The LMC was also designed to monitor local joint temperature, torque, position, current, and rotor position sensors for motor commutation. Custom schematic design and multilayer board fabrication allowed direct integration within the drive module.
This approach allowed the drives to be designed as single integrated motor and controller packages, helping to shrink the overall mechanical profile and maximizing performance. Each LMC uses an advanced reduced instruction set computer (ARM)-based processor.
The researchers wanted to create prosthesis capable of extremely fine dexterity and precision, allowing users to perform tasks ranging from the mundane, such as turning a doorknob, up to and including playing a piano.
“We want people to be able to do the very complex things with their fingers that those of us who have an arm and hand do naturally,” McLoughlin said.
A biomechatronic ankle
Perhaps the most impressive biomechatronic prosthesis currently available is the BiOM T2 ankle-foot prosthesis, which allows users with below knee amputations to walk naturally and, in many cases, with no trace of a limp.
The BiOM T2 was developed at MIT’s Biomechatronics Lab, led by Hugh Herr — himself a double below-knee amputee. It differs from conventional leg prostheses by emulating muscle function rather than relying on the user’s remaining muscles to provide motion energy.
Herr has compared the robot ankle to a car that uses an engine and powertrain to transport its occupants as opposed to a bicycle that requires its rider to supply all the energy.
The BiOM T2, which is owner-programmable, uses sensors and microprocessors to control a carbon spring that functions like a human foot. The system, exclusive of the modular battery, is designed to achieve a lifespan of three to five years of ordinary use, which equates to about five million steps by a 250-pound user.
The battery, which needs about 45 minutes to charge, can supply power to the system for about four to six hours.
With a price tag of over $50,000, the BiOM T2 costs about two to three times as much as a conventional artificial leg.
On the other hand, the self-powered device allows users to walk faster and with a more natural stride, which reduces physical stress and lowers the risk of complications, such as knee osteoarthritis.
BionX Medical Technologies Inc., the Bedford, Mass.-based company that makes and markets BiOM, notes that the device also increases its user’s stair-climbing “push-off power” to the level of that of non-amputees.
A new sensation
Veronica Santos, an associate professor of mechanical and aerospace engineering at the University of California, Los Angeles, and Jonathan Posner, an associate professor of mechanical engineering at the University of Washington in Seattle, are developing a multimodal tactile sensor skin that promises to reduce the cognitive burden on prosthetic hand users, making control a faster, intuitive, and more natural process.
A current prototype can detect various finger-object interactions, such as normal contact forces and low-frequency dynamic loads.
Its pliable, sensor-laden skin, which is closely wrapped around a prosthetic hand’s curved digits, aims to help users grip objects by cushioning impacts, increasing the effective contact area during grasp and enabling activities of daily living that rely upon a sense of touch.
“We are developing algorithms to map artificial tactile sensor data to object properties for restoring the sense of touch to amputees,” Santos says.
The system’s capacitive sensors are created by injecting a flexible material, such as a polydimethyl siloxane (PDMS) polymer, with a liquid metal alloy that’s designed to function as deformable wires and plates.
“If we monitor the voltage across those plates, we can relate that data to the distance between the plates, which we can then relate to the deformation of the skin surface,” Santos says. “The raw voltages can be turned into something representing a contact area, an indication of force being applied, or the hardness of an object, for example.”
Such information can be relayed to amputees via neural interfaces or used to develop intelligent artificial reflexes and grasp patterns within the prosthesis itself.
Ready for the real world?
As new biomechatronic prostheses are developed, many researchers wonder how well these complex devices will adapt to the rough and tumble world outside of the lab. The BiOM T2, for instance, has a projected lifespan of only three to five years of ordinary use.
“There are still challenges,” Santos says. “The technology’s drawback is that I can do something really cool on the lab bench, but if I took that device and strapped it to someone and sent them home, it wouldn’t last more than two minutes.”
Despite the challenges, many researchers believe that biomechatronic systems are an important new technology in the form of replacement limbs as well as wearable devices designed to enhance the performance of healthy arms and legs.
Steve Collins, director of the Experimental Biomechatronics Laboratory at Pittsburgh’s Carnegie Mellon University, believes that biomechatronic devices could eventually weave their way into everyday life.
“If we make the benefits great enough, in 10 or 20 years, people will use them recreationally, perhaps just to make it easier to walk an extra few blocks,” he predicted.