Analysis and Considerations of the Controllability of EMG-Based Force Input

Using electromyography (EMG) measurements for user interfaces (UIs) is widely employed as an interaction method. Some advantages of using EMG-based input are that it does not require a physical controller and can be operated intuitively with small body movements. Existing work has explored different novel interaction methods for UIs using EMG. However, it is still unclear how precisely users can control the force and what kind of control pattern is easier for them to use. Thus, this paper analyzes the effect of EMG-based force input on control accuracy and mental workload. We constructed a pointer-tracking application that inputs force strength from forearm EMG. Tracking accuracy and mental workload were evaluated under the conditions of multiple tracking patterns and hand gestures. The results showed that EMG-based input accuracy was affected by the way in which the force was applied (e.g., strengthened, weakened, or fluctuated). We also found that hand gesture type did not influence accuracy or mental workload.