American engineers have developed a new control algorithm for active
prosthetic leg above the knee, allowing the wearer to cross
obstacles. The algorithm tracks the movement of the stump, and at every step
adjusts the trajectory of the prosthesis, including stronger leg compresses, if
there is a need, say the authors of the article in Science is Robotics.
There are two main types of amputation of a leg, and they are highly dependent on how comfortable the person will be wearing a prosthesis. If the leg is amputated below the knee, the person can control the movements, bending the knee and prosthesis remains only to control the bending angle of the ankle joint. If levels of amputation above the knee, you need a much more complex prosthesis and the control algorithm to them, he must lead angles at the two joints and is responsible for a large part of the legs. Usually people with a leg amputated above the knee, using semi-active prostheses that can not fully compensate for the function of the knee and in particular do not it is easy to step over obstacles.
Lenzi Tommaso (Tommaso Lenzi) and his colleagues from the University of Utah created an algorithm for the active above-knee prostheses, allowing them to recognize that man wants to lift his leg to step over an obstacle. Engineers used a prosthesis, developed by them earlier in the previous work, and focused on the control algorithm of its work.
The developers decided essentially to shift some of the control the prosthesis in a person. When a person is faced with an obstacle, after which it can cross, he’s stronger, pushes the thigh and lower leg to each other during a step farther and brings the thigh forward. Accordingly, by measuring the movement of a hip, the prosthesis can recognize the tilt angle exceeding a threshold value, and understand that the engine in the knee needs more press galanou part to the femur.
Engineers have developed a simple algorithm that dynamically, at each step, collects from the accelerometer settings foot movements, including position, acceleration and speed, it updates the planned motion path of the prosthesis, and gives the motor at the knee of the low-level commands. Tests of the prosthesis in three volunteers showed that they are able to arbitrarily control the movement of the prosthesis and to step over obstacles of different heights up to 20 inches. The authors note that will continue to be tested on a larger sample and directly compare the performance of the new algorithm with passive prostheses and active, running other algorithms.
Last year, American engineers taught the prosthetic leg with a knee quickly, in about ten minutes, to adjust the parameters of walking by a particular person by analyzing the motion parameters of the other leg. Similar algorithms adapting to the gait of a particular person uses active exoskeletons legs. This allows them after a short calibration to work more effectively and significantly reduce energy costs of human walking or running.