Design Engineering

Best Foot Forward

By Treena Hein   

General design prosthetics Robotics

Robotic POWER KNEE prosthesis restores natural movement to lower-limb amputees

For every would be engineer who grew up watching The Six Million Dollar Man, the idea of electro-mechanical implants or prosthetics that recreate the function of missing limbs remained a daydream. For one of them, however, it became a profession.

In 1998, Stéphane Bédard, at the time a mechanical engineering graduate student at Laval University, had just returned from competing in mountain bike racing at the national level. Injured, he found himself once again in a rehabilitation centre following his third knee surgery. As he looked around, day after day, at the other patients, inspiration struck.


“I noticed during my three stays in rehab that there weren’t a lot of new technologies being used for amputee patients,” he says. “The apparatus were very primitive. I wanted to help these patients walk better and just feel better. I had a lot of skill and knowledge in robotics and automation and knew I could be the one to do this.”

The POWER KNEE, the first commercial lower-limb prosthesis that is both motorized and processor controlled, allows amputees to walk naturally and perform common tasks like climbing stairs.

Bédard says he essentially left everything behind and founded Victhom Human Bionics, a medical device company located in Saint-Augustin-de-Desmaures, Que., that specializes in the development and commercialization of bionic devices. The company’s signature product is a motorized, processor-controlled lower-limb prosthesis for people amputated above the knee.

Manufactured under the name POWER KNEE by Ossur, an Icelandic company specializing in non-invasive orthopedics, Victhom’s “bionic leg” is the first commercially available prosthetic knee to provide the power necessary to restore amputated lower-limb function and allow its users to not only walk without assistance but also perform common tasks. “It’s an active prosthesis that allows a person to walk and use stairs naturally,” Bédard says. “Others are not motorized and people use their own energy to move the prosthesis.”

Development of the POWER KNEE began in the 1990s and progressed rapidly. By 2000, Bédard had his design complete and, by mid-2001, had hired his first employee. “I needed a lot of money at that point,” he says. “For the first financial round, I received $1.1 million from Innovatech Quebec, an early-stage venture capital fund owned by the Government of Quebec. All the rest of the three million in financing was contributed by private investors.” In all, he says it took about 40 scientists and engineers and many millions of dollars to develop the first functional prototype by 2004 and get the first POWER KNEE to market in 2006.

Design details
While the POWER KNEE wasn’t the first processor-controlled prosthetic leg, its motorized knee joint and sensor system design do set it apart from other similar prosthetics available. For instance, Otto Bock HealthCare’s C-Leg uses hydraulic cylinders in the knee joint to control the amount of flex in the knee joint. Force sensors in the prosthetic sample heel, toe and axial loading data 50 times per second and feed that it to a microprocessor that adjusts lithium-battery powered valves that adjust the resistance of the joint.

By contrast, the POWER KNEE locates its kinetic and kinematic sensors inside the shoe of the amputee’s sound leg (the non-amputated leg), which are wired to a transmitter module on the ankle. This information is then transmitted by radio frequency (Bluetooth) technology to a DSP-based controller within the prosthesis.
The on-board artificial intelligence module interprets the information and then sends commands to a motor in the prosthesis’ knee joint to generate the required trajectory and perform common tasks, such as walking on uneven ground or climbing stairs. In this way, Bédard says, the artificial leg can anticipate and mirror, step by step, the speed and activity of the sound leg.

“All the methodology is based on fundamental concepts we can find in the field of robotics,” he says. “What’s different is the type of sensors we are using to model human locomotion and our proprietary artificial intelligence software, which translates information from the sound leg into motor control.”

Bédard says developing a processor-driven system that could synchronize the robotic artificial leg with the movement of a user’s sound leg was relatively easy. The real challenge came in getting the POWER KNEE’s artificial intelligence to recognize and respond when a user wants to change from one form of locomotion to another.

“Let’s say you want to go from walking to climbing stairs, that transition must be recognized in real time,” he explains. “These types of common in-between locomotion mode shifts are very important and they are the main things to optimize when you want to create a commercial product.”

In addition to the software, another major challenge was integrating all the necessary components into the prosthesis itself while providing sufficient torque at the knee joint and enough battery life to last for approximately a day’s worth of use. To provide the power, the robotic leg uses a standard brushless DC motor. Its rotational motion, Bédard says, is changed into linear motion with an endless screw, which connects to the patented mechanical design of the knee joint’s frame to create angular motion.

“The selection of a DC motor was necessary to provide the torque needed while still being small enough to fit within the space available,” he says. “For the power supply, we designed our own and used off-the-shelf rechargable lithium polymer batteries, but we arranged them so that the leg would allow a 250-pound person to climb 100 steps in a row. The important thing was to design the prosthesis so that everything—the motor, sensors, power supply, motherboard and batteries—would all fit within the prosthesis and deliver a purely autonomous product.”

Ahead of its time
Although the POWER KNEE is revolutionary, only a few units have been sold worldwide so far. “There is no historical precedent for this using this type of innovation,” Bédard says. “It wasn’t very well received at the beginning. We had to educate the market about the clinical benefits of this new type of prosthesis.”
In addition, he says it’s the most expensive prosthesis on the market and has technical limitations in terms of its size, weight and functionality. To address these drawbacks, the company is currently working on a second version that will accommodate a wider audience.

“The first version is for active and very active people,” Bédard says. “This second version will be better able to meet the needs of people with less energy, including those that are heavier, elderly and/or battling disease. To do that, we are cutting its weight and noise and making it shorter.” It will also be able to be fitted to shorter people and teens.

He sees the future of the prosthesis moving toward more transparency for the user. Eventually, he says, Victhom will develop technology that allows a person’s central nervous system to control or complement controller function.

“We would also like to improve the socket that the amputee’s stump fits into,” he adds. “It is now custom molded, but if we are able to render this socket active and dynamic, we can change its shape and provide more control of the prosthesis.”

Treena Hein ( is a Pembroke, Ont.-based freelance writer.


Stories continue below

Print this page

Related Stories