Natural gait event-based level walking assistance with a robotic hip exoskeleton

Area of Science:

• Robotics
• Biomechanics
• Human-Computer Interaction

Background:

• Robotic exoskeletons offer potential for gait assistance.
• Effective assistance requires precise timing and user-specific adaptation.
• Current strategies often lack adaptability to individual needs.

Purpose of the Study:

• To present a novel assistance strategy for level walking using a robotic hip exoskeleton.
• To enable systematic adjustment of assistance levels based on wearer preference.
• To define clear principles for initiating and terminating robotic assistance during gait.

Main Methods:

• Utilizing an inertial measurement unit (IMU) on the pelvis to estimate foot contact events.
• Defining gait cycle phases based on reliable sensor-detected natural gait events.
• Pre-determining maximum torque, duration, and torque profile for assistance at the start of each step.
• Conducting experiments on a motorized treadmill to assess metabolic cost.

Main Results:

• The strategy successfully utilizes IMU-based foot contact detection for gait phase transitions.
• Assistance quantity can be systematically adjusted to wearer needs and preferences.
• The system provides explicit initiation and termination of assistance.
• Experimental data on metabolic cost during treadmill walking is collected.

Conclusions:

• The presented strategy offers a reliable method for robotic hip exoskeleton assistance during level walking.
• The approach allows for personalized and adaptive gait support.
• Further investigation into metabolic cost reduction is warranted.