Enhancing human mobility using robotics, wearables, and AI

Persons with lower-limb disabilities experience significantly limited mobility due to neurological damage or advanced age. Wearable assistive devices, such as robotic exoskeletons, have the potential to transform human mobility by restoring motor functions for these individuals and thereby providing independence. My research program undertakes the grand challenge of enhancing human mobility by optimizing wearable robot control frameworks that integrate robust mechatronic designs, advanced artificial intelligence, and an understanding of human motor behavior. This broad research spectrum combined with a multifaceted experimental approach is unique and enables a multi-perspective view of human-machine systems, facilitating the creation of next-generation robotic systems. Significant mobility improvements and retention of motor functions in the clinical population are only possible with an intelligent and intuitive exoskeleton system that seamlessly adapts to the user, understands how humans learn to use these devices, and can rapidly tune to different populations. My research serves as a foundation for a broader movement to take wearable robots into the real world making a significant societal impact.

Robotic exoskeletons for dynamic human locomotion

A detailed understanding of human gait biomechanics is required to design advanced exoskeletons that can effectively improve human performance. However, the interdisciplinary nature of wearable systems requires a significant amount of knowledge in multiple domains, rendering the entire design process non-trivial. To date, I have developed several unique autonomous high-fidelity robotic exoskeletons, capable of assisting a diverse set of locomotor tasks. These exoskeleton designs have been optimized to provide joint assistance comparable to human biological joint moments while minimizing the system’s overall mass and accounting for user comfort. The designs match the current state-of-the-art hip exoskeleton design in key areas such as actuator design and user interface. Furthermore, these systems can assist with a significant amount of joint torques and velocity, which allow the robot to operate during intensive ambulation scenarios such as ramp and stair ascent or highly agile movements. Moreover, the system has an integrated AI co-processor, allowing the exoskeleton to operate under a multi-tier control scheme. This highly effective control framework allows the entire system to perform real-time inference of the user’s state and accurately provide a biomechanically optimal level of assistance during locomotion.

Previous robotic hip exoskeleton designs

Relevant publications

Kang et al., Design and validation of a torque-controllable series elastic actuator-based hip exoskeleton for dynamic locomotion, Journal of Mechanisms and Robotics 2023
Kang et al., Electromyography (EMG) signal contributions in speed and slope estimation using robotic exoskeletons, IEEE International Conference on Rehabilitation Robotics (ICORR) 2019
Kang et al., The effect of hip assistance levels on human energetic cost using robotic hip exoskeletons, IEEE Robotics and Automation Letters 2019