Skip to main content
Figure 3 | Journal of NeuroEngineering and Rehabilitation

Figure 3

From: Control of thumb force using surface functional electrical stimulation and muscle load sharing

Figure 3

Controller scheme. Block diagram of feedforward and feedback thumb force controller. Stimulation for three individual muscles is calculated based on a reference force. Force distribution over the muscles is calculated by solving a QP problem as shown in 5 indicated by the 'QP’ blocks. These QP solvers take the previously determined force map and also boundaries on the recruitment into account. For clarity this is left out in the schematic. The bounds for the feedforward QP problem are [0,1]. The bounds for the feedback QP problem depend on the current activation of the muscle (from both feedforward and feedback path) and indicate the remainder of the operating range ([0,1]) and can thus also be negative when the specific muscle is already active. In the feedback path a PI controller was used for each individual muscle force. When using a combination of feedforward and feedback control, the feedforward path was reduced by a factor K=0.8 to prevent overshoot and let the feedback path compensate for the remainder. When evaluating the feedforward control performance without feedback, K was set to 1.

Back to article page