Developing safe fall strategies for lower limb exoskeletons.

Area of Science:

• Robotics
• Biomechanics
• Human-Robot Interaction

Background:

• Powered lower limb exoskeletons (LLEs) present balance maintenance challenges, posing safety risks like head or hip injuries during falls.
• Current LLE safety limitations restrict their community use to supervised settings.
• Existing control strategies do not prevent falls, necessitating research into safe fall mitigation.

Purpose of the Study:

• To develop safe fall control strategies for LLEs using optimization techniques.
• To minimize injury severity by avoiding head impact and reducing hip impact velocity during simulated falls.
• To investigate optimal joint trajectories and torque profiles for LLEs during a fall.

Main Methods:

• Developed an optimization methodology based on human biomechanics data to model backward falls.
• Validated the methodology by comparing simulated fall kinematics and dynamics to real-life human falls.
• Extended the methodology to incorporate LLE characteristics, generating optimal control strategies for fall mitigation.

Main Results:

• The optimization methodology accurately replicated kinematic and dynamic characteristics of human falls.
• Simulated LLE falls using optimized strategies avoided head impact.
• Optimized fall strategies reduced hip impact velocity by over 50% compared to LLEs with locked joints.

Conclusions:

• Optimized control strategies can significantly enhance LLE safety during falls.
• This approach mitigates injury risk, paving the way for safer independent LLE use.
• Further research can integrate these strategies into LLE control systems for community deployment.