Toward Remapping Residual Movement of Shoulder: A Soft Body-Machine Interface

Spinal cord injury-induced tetraplegia significantly impairs patients’ ability to carry out daily activities. Customized human-machine interfaces for controlling assistive devices is vital in improving the patient’s self-help ability and reducing the societal caregiving burden. In this study, we developed a body-machine interface that utilizes soft sensors and IMUs to remap residual shoulder movements of tetraplegics to a two-dimensional continuous command space. A rule-based and a user intent-inference-based data decoding method were designed by us, respectively, for mapping interface data to command spaces. By involving the prior knowledge of user’s abilities into a shared autonomy framework, we have implemented a shared command mapping approach to enhance the efficiency of command generation. Finally, we validated the effectiveness of the proposed method through target reaching tasks and a virtual powered wheelchair driving task. Our proposed method has the potential to be integrated into clothing, enabling non-invasive interaction in daily life. Additionally, it can also serve as a complementary for completing intricate tasks through cooperation with brain-computer interface or joysticks.