Coordinated human-exoskeleton locomotion emerges from regulating virtual energy

Abstract

Lower-limb exoskeletons have demonstrated great potential for gait rehabilitation in individuals with motor impairments; however, maintaining human-exoskeleton coordination remains a challenge. The coordination problem, referred to as any mismatch or asynchrony between the user's intended trajectories and exoskeleton desired trajectories, leads to sub-optimal gait performance, particularly for individuals with residual motor ability. Here, we investigate the virtual energy regulator (VER)'s ability to generate coordinated locomotion in lower limb exoskeleton. Contribution: (1) In this paper, we experimented VER on a group of nine healthy individuals at different speeds (0.6m/s - 0.85m/s) to study the resultant gait coordination and naturalness on a large group of users. (2) The resultant assisted gait is compared to the natural and passive (zero-torque exoskeleton) walking conditions in terms of muscle activities, kinematic, spatiotemporal and kinetic measures, and questionnaires. (3) Moreover, we presented the VER's convergence proof considering the user contribution to the gait and introduced a metric to measure the user's contribution to gait. (4) We also compared VER performance with the phase-based path controller in terms of muscle effort reduction and joint kinematics using three able-bodied individuals. Results: (1) The results from the VER demonstrate the emergence of natural, coordinated locomotion, resulting in an average muscle effort reduction ranging from 13.1% to 17.7% at different speeds compared to passive walking. (2) The results from VER revealed improvements in all indicators towards natural gait when compared to walking with a zero-torque exoskeleton, for instance, an enhancement in average knee extension ranging from 3.9 to 4.1 degrees. All indicators suggest that the VER preserves natural gait variability and user engagement in locomotion control. (3) Using VER also yields in 13.9%, 15.1%, and 7.0% average muscle effort reduction when compared to the phase-based path controller. (4) Finally, using our proposed metric, we demonstrated that the resultant locomotion limit cycle is a linear combination of human-intended limit cycle and the VER's limit cycle. These findings may have implications for understanding how the central nervous system controls our locomotion.

Fig 1. The conceptual difference between conventional time-based controllers and virtual energy regulator.

Human-exoskeleton system dynamics at joint level can be expressed in the time-position-velocity space, two of these dimensions are sufficient to design a controller for such systems. Accordingly, three different scenarios can be imagined: i) position-time resulting in trajectory tracking controllers, ii) velocity-time resulting in velocity-based controllers, and iii) position-velocity resulting in limit cycle control (virtual energy regulator). The time-based controllers (i and ii) are sensitive to time offset between human intended and exoskeleton desired trajectories as they misinterpret a time difference as kinematic error (left figure) which can reduce human-exoskeleton coordination. In contrast, controlling in position-velocity domain (or virtual energy domain), allows time-independence and robustness to the time offsets, resulting in a better human-exoskeleton coordination (right figure).

Fig 2. Desired limit cycles and VER applied torque and power illustration.

(a-b) compare the natural and designed trajectories at the knee and hip joints, where the gray background indicates the stance phase [42]. (c-d) compare the natural and desired limit cycles at 0.85m/s [42]. (e-h) illustrate VER applied torque and power w.r.t. the desired limit cycle. The color map shows the torque distribution w.r.t. the designed limit-cycles; yellow(dark blue) is positive(negative) highest value.

Fig 3. The experimental setups.

The experimental setups, including Indego explorer exoskeleton (Indego, Parker, USA), 16 surface electromyography (sEMG) Trigno sensors (Delsys, USA), split-belt instrumented treadmill (Bertec, USA) equipped with two 6-degree of freedom (DoF) force plates, and optoelectronic motion capture system containing eight Vero 2.2 cameras (Vicon, Motion System, UK) 16 reflective markers are placed according to the “PlugIn-Gait” recommendation for lower body kinematic measurements.

Fig 4. Muscle activation pattern comparison at 0.85m/s.

Comparison between muscle activation patterns and average muscle activity for a representative participant in four different conditions; Natural, Passive before, Active, and Passive post. For this representative participant, VER (Active condition) results in average muscle effort reduction of 17.5% compared to Passive condition.

Fig 5. The kinematic comparison at 0.85m/s.

(a,b,e,f) compare the Desired trajectory and limit cycle with Passive, Active, and Natural conditions at hip and knee joints for a representative participant. (c,d,g,h) compare the correlation coefficient and RMS of deviations from the Desired and Natural trajectories with Active and Passive conditions at hip and knee joints across all participants. (i) shows the toe clearance trajectory for a representative participant, and (j) compares minimum toe clearance of Passive, Active, and Natural conditions across all participants. (k,l) compare Passive and Active conditions in terms of hip(knee) maximum swing(stance) flexion(extension) across all participants; zero angle corresponds to a fully extended knee.

Fig 6. VER control performance at 0.85m/s.

(a,b) describe virtual energy against the limit cycle phase where the gray backgrounds indicate the stance phase. (a,b) compare the desired virtual energy with virtual energy in two different conditions (Passive and Active) at hip and knee joints for a representative participant. (c,d) compare Pearson correlation coefficient and RMS of deviation from the desired virtual energy in two different conditions (Passive and Active) for hip and knee joints across all participants. (e,f) The questionnaire results for all participants at 0.6m/s and 0.85m/s. The box plots compare Passive with Active condition in terms of comfort, safety, stability, effort, and time to fatigue. The vertical axes for time to fatigue is in right side of the plots.

Fig 7. Comparison between VER and PBP controllers at 0.8m/s.

(a) illustrates the muscle effort reduction (of eight different muscles) due to VER compared to the PBP case for three different participants; the positive values indicate VER-resultant relative muscle effort reduction. (b-e) compare the participants knee and hip joint kinematic when using VER and PBP controllers.