Assistive Exoskeleton Control with User-Tuned Multi-Objective Optimization

Assistive exoskeletons that utilize trajectory following control have been shown to produce stable gait for users. These however, do not allow intuitive tuning to customize gait to users' preferences. When persons walk on their own, they balance a variety of needs such as speed, comfort, and energy. Providing user tuning by optimizing between different gait performance measures gives an intuitive flexibility. We have shown the optimization between natural walking and gait energy produces stable bipedal gait through simulation in a virtual constraint framework. This verification shows validity of the methodology and framework for improving tuning and customization of assistive exoskeletons.