Optimized Control of Different Actuation Strategies for FES and Orthosis Aided Gait

A combination of functional electrical stimulation (FES) and an orthosis can be used to restore lower limb function in persons with paraplegia. This artificial intervention may allow them to regain the ability to walk again, however, only for short time durations. To improve the time duration of hybrid (FES and orthosis) gait, the muscle fatigue due to FES and the fatigue in arms, caused by a user’s supported weight on a walker, needs to be minimized. In this paper, we show that dynamic optimization can be used to compute stimulation/torque profiles and their corresponding joint angle trajectories which minimize electrical stimulation and walker push or pull forces. Importantly, the computation of these optimal stimulation or torque profiles did not require a predefined or a nominal gait trajectory (i.e., a tracking control problem was not solved). Rather the trajectories were computed based only on pre-defined end-points. For optimization we utilized the recently developed three-link dynamic walking model, which includes both single and double support phases and muscle dynamics. Moreover, different optimal actuation strategies for FES and orthosis aided gait under various scenarios (e.g., use of a powered or an unpowered orthosis combined with stimulation of all or few selected lower-limb muscles) were calculated. The qualitative comparison of these results depict the advantages and disadvantages of each actuation strategy. The computed optimal FES/orthosis aided gait were also compared with able-bodied trajectories to illustrate how they differed from able-bodied walking.