Control Paradigm for Unilateral Robotic Exoskeletons to Assist Hemiparetic Gait

The assistance provided by a robotic exoskeleton would be fully effective when this device is embodied by the user, so the robot becomes a part of the own user’s body image instead of an external tool. This embodiment implies that the user can predict the action of the robot, and it matches their expectation. This issue is especially crucial for partial robotic exoskeletons that do not act over the complete user’s locomotor system, but they act only partially on some joints, so these devices should be able to synchronize with the user’s residual gait. This chapter presents a control paradigm for unilateral robotic exoskeletons to assist post-stroke hemiparetic gait that coordinates its action with the actual gait of its user. This coordination relies on the movement of the non-assisted leg to estimate the continuous gait phase through an adaptive frequency oscillator. Based on this estimation, the device synchronically generates an assistive pattern following two different control strategies: (1) replicating the movement of the non-assisted leg or (2) synchronizing a healthy gait pattern. This chapter also includes the validation of the generated assistance with experimental data of three healthy subjects and a brief discussion about the feasibility of the generated patterns.