abstract
- © 2020, Springer Nature Switzerland AG.Three-dimensional printed prostheses have gain popularity due to their cost, customization capabilities and rapid prototyping, which allows researchers to implement over them various control and classification techniques that make them more functional for the users. This paper presents the design, construction and instrumentation of an electromyographic (EMG) active upper limb prosthesis, with a set of distributed PD controllers that ensures tracking of trajectories corresponding to different arm movements. The proposed rehabilitation device employs an artificial neural network to exert the classification of EMG signals which drive the activity of the prosthesis. The classifier defines the reference trajectories which must be tracked by the PD controllers. Finally, the complete integration of the system is presented, as well as the results obtained by each of its parts.