Abstract
Individuals aged 65 and above face a heightened risk of falls, which can lead to severe and potentially life-altering consequences. Assistive technologies that support the elderly in their daily activities while reducing fall risk are crucial for enhancing both longevity and quality of life. Despite the established body of research on motion-assistance exoskeletons, there remains a significant gap in our understanding of robot/user interaction leading to limited functionality and usability of balance-support devices. Thus, this research aimed to advance the current fall prevention technology through the development and evaluation of an innovative wearable robot designed for balance assistance.The robot was a powered wearable lower limb exoskeleton designed to adjust stride length to improve obstacle avoidance during walking. The robot was able to control the foot position in the x-direction with only one active degree of freedom per leg located at the hip joint. Its non-anthropomorphic structure allowed for distinctive features such as a telescopic leg to directly connect the active hip joint to the target ankle joint and a compact between-leg configuration.
Experiments involving human participants were conducted in the Human-Environment Advanced Dynamics laboratory at the University of Bath (REACH Ref: EP 22 075). Three participants took part in walking trials on a treadmill whilst wearing the developed robot. Their movements were recorded simultaneously using both the instrumented robot and an independent motion capture system. The robot randomly applied different test conditions to steps to test the effect of torque magnitude, direction, and profile shape.
The robot was able to apply a force to the swing foot to successfully alter the user’s stride length, indicating the possibility for obstacle navigation assistance. Positive torques resulted in longer strides whereas negative torques resulted in shorter strides. However, the effects of corrective movements by exoskeletons to support balance have been known to inadvertently disrupt the user’s natural balance. Thus, it was important to investigate the wider effects to the user’s biomechanics as a response to the robot’s actuation.
These novel experiments looked at systematically analysing the user’s response to corrective actions by the robot. It was clear from the kinematic changes to both of the user’s lower limbs (test and contralateral leg) and centre of mass that the ultimate change in stride length was related to the user’s response to the actuation. Negative torques resulted in larger kinematic changes although this was not initially evident in the stride length changes. There were also distinct differences between subjects which may indicate different strategies of resistance/compliance to the manipulation of the stride length, potentially associated with leg elevating and lowering strategies seen in trip recovery. In addition to these findings, a lower torque development rate was associated with potentially less user resistance or a lower degree of robot perturbation. A deeper insight into how the nature of the torque application influences user response may prove crucial for the enhancement of subsequent controller designs.
These findings provide insights into the importance of understanding robot/user interaction and the implication for the development of improved fall prevention devices, paving the way for enhanced balance assistance devices for the elderly population.
| Date of Award | 11 Sept 2024 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Erfan Shahabpoor Ardakani (Supervisor) & Andrew Plummer (Supervisor) |