# Active robotic assistance for standing and sitting: experimental evaluation of handle trajectories

## TL;DR

This paper evaluates robotic handle movements to help people stand up and sit down, reducing leg strain but increasing upper body effort.

## Contribution

The study introduces and evaluates four scalable handle trajectories for robotic assistive devices to reduce leg loading during standing and sitting.

## Key findings

- Proposed handle trajectories reduced peak hip and knee extension moments by over 70% and 50%, respectively.
- Vertical handle forces increased by over 30%, reaching up to 60% of body weight.
- Participants perceived reduced lower limb load but increased upper body demand with moving handles.

## Abstract

Standing up and sitting down are important activities of daily living, but require large leg moments that often exceed the muscle strength of older adults. Some robotic rollators are designed to provide standing-up and sitting-down assistance through actuated handles or armrests to reduce the loads on the legs, but it is still unclear how they should move. There is limited information on appropriate assistance trajectories and their effects on the body during standing up and sitting down.

We designed four physiological, scalable and parameterized handle trajectories based on unassisted shoulder movement that can be readily implemented in robotic assistive devices, and evaluated their effect on leg loading, energy input, handle forces and perceived assistance in 15 healthy younger adults. We created a robotic assistance simulator device equipped with moving handles to compare the trajectories to static handles (representing a conventional rollator), and collected full-body motion, ground reaction forces, handle forces and scored perceived assistance.

The proposed handle trajectories substantially decreased leg loads compared to the static handle assistance (non-moving handle), with the two best-performing trajectories reducing the peak hip extension moment by over 70% and the peak knee extension moment by over 50% during standing up and sitting down. This is associated with an increase in peak vertical handle forces of over 30%, with the total bilateral vertical forces reaching up to 60% of body weight, and a decrease in peak horizontal force of more than 50%. The subjective participants’ perception reflected the lower limb mechanical load. The handle velocity was shown to play a secondary role within the investigated range.

The proposed support trajectories can be scaled to the person’s anthropometry and readily implemented in robotic assistive devices, and were shown to substantially reduce leg loading, potentially improving life quality of individuals with difficulties in standing up. However, the large vertical handle forces and thus upper body demand during moving-handle assistance is a trade-off with relieving the lower limb load. This work provides a comprehensive foundation for the design of the necessary further experimental assessments with the target population.