The more complex and functional a prosthesis or implant for a disabled person, the longer and more difficult it is to customize it for a specific person. Sometimes it takes weeks of practice. New technology based on artificial intelligence will solve this problem in almost one session.
The technology is experimental, somewhat controversial, and therefore its developers do not want undue attention yet. The highlight of the method is that the AI inside the prosthesis is trained, including on the errors of its user. Which in the case of a robotic knee means a violation of the rhythm of walking until a person falls. It is physically painful and is somewhat fraught with new injuries - but without bumping, you cannot achieve a real result.
The robotic knee uses an algorithm to analyze data from sensors that must fluctuate within specified limits, which corresponds to a calm measured walk. If the indicators go beyond the boundaries, the AI adjusts the movement mode and makes adjustments to the operation of the prosthesis nodes. On average, it takes a few minutes to adapt to a new user.
Alas, the technology also lacks shortcomings at this stage. Firstly, the device is energy-intensive, and the AI itself is located in an external module - you have to move with bundles of wires on your leg. Secondly, he knows how to work only when driving on a flat straight road; the system is not yet able to master maneuvering. Thirdly, the prototype is not universal and not suitable for all patients. However, all these problems can be solved - and it is possible that this development may become revolutionary in the creation of new prostheses and implants.