← Paper Digest

Levio: Smart Actuated Back-Exoskeleton via Retrofit of a Commercial Passive Device

Ionnais Kyriazis, Mohamed Irfan Refai, Herman van der Kooij

November 2026

TL;DR They retrofit a commercial passive back exoskeleton (Laevo Flex) with two off-the-shelf quasi-direct-drive motors, IMUs and a load cell, and switch its assistance between the bending and lifting phases. In 10 kg stoop lifts, it lowered resistive torque while bending from 25.2 to 17.9 Nm and raised lifting assistance from 30.6 to 37.7 Nm compared with the passive device.

What they do

Passive back-support exoskeletons give a fixed, spring-defined assistance that peaks while the user bends rather than when they lift. Building an active one from scratch means redesigning the human interface. Levio keeps the frame, vest and thigh interfaces of a commercial passive device (Laevo FLEX V3.0) and swaps its gas springs for motors, so the original device serves as a direct baseline.

How it works

  • Actuation. One CubeMars AK10-9 quasi-direct-drive motor per side, driven by MAB MD80 controllers and sized for 40 Nm per side at 120°/s. A passive alignment hinge keeps the original stance-width freedom.
  • Sensing. IMUs on torso and pelvis give the torso-pelvis angle. A rear load cell measures the assistance actually transmitted through the vest.
  • Control. A Raspberry Pi runs a ~50 Hz state machine: a bending profile at 60% of nominal torque, then a latched lifting profile with an extra boost once angle and angular-velocity conditions trigger. Commands are low-pass filtered and slew-rate limited, and a hardware e-stop cuts power.

Results

  • Thigh connectors deformed at 40 Nm, so tests were capped at 25 Nm per side.
  • Nine participants stoop-lifted a 10 kg disc with the passive Laevo, Levio at zero torque, and Levio assisting. Kinematics did not noticeably change.
  • Mean interface torque (seven participants): bending 25.2 ± 4.3 Nm passive vs 17.9 ± 4.6 Nm active; lifting 30.6 ± 9.5 vs 37.7 ± 6.4 Nm. Zero-torque residual was 1.9-2.7 Nm RMS.
  • Usability score 81.7 active vs 76.9 passive and 65.3 zero-torque; global discomfort 2.6 vs 3.6 and 3.8, with the largest drop at the lower back.

Limitations

Mass rose from 4.3 to about 7 kg, and the 40 Nm target went unverified. Only stoop lifting was evaluated: squat trials showed thigh-pad migration, and without a thigh IMU the measured angle no longer matched the exoskeleton’s. Small sample, descriptive statistics only, and the load cell measures interface force, not lumbar unloading.

Figures

From Figure 4: Back and side views of Levio. The actuator sits between the torso-side force-transfer bars and the thi...
From Figure 4: Back and side views of Levio. The actuator sits between the torso-side force-transfer bars and the thigh-side structure of the Laevo frame, and a separate alignment hinge allows out-of-plane leg motion. The rightmost panel shows the final version with machined parts.
From Figure 1: Exploded view of the mechanical parts that integrate the actuator into the frame. The alignment hinge ...
From Figure 1: Exploded view of the mechanical parts that integrate the actuator into the frame. The alignment hinge preserves the passive degrees of freedom.
From Figure 3: Target bending and lifting assistance torques as a function of torso-pelvis angle.
From Figure 3: Target bending and lifting assistance torques as a function of torso-pelvis angle.
From Figure 6: Average estimated assistive torque over the lift cycle for the passive, zero-torque and active assista...
From Figure 6: Average estimated assistive torque over the lift cycle for the passive, zero-torque and active assistance conditions. Shading is the standard deviation across participants.
Abstract

Passive back exoskeletons show promising benefits for reducing occupational musculoskeletal disorders. However, their assistance cannot be modulated towards the task requirements. Active exoskeletons offer opportunities for pose-specific assistance, but, developing these systems from scratch for research can be time consuming. Therefore, here, we present an accelerated design pipeline that retrofits a Commercial Off-The-Shelf (COTS) passive back exoskeleton, Laevo Flex, with COTS actuation (CubeMars AK10-9) and sensing (IMUs and load-cell) paradigms. Levio was able to provide assistance based on the user's pose. Nine participants performed stoop lifting with the original Laevo and the retrofit Levio system, in transparent and assisted modes. Compared to Laevo, Levio modulated the assistance provided between the bending and lifting phases of the stoop task. The passive assistance offered by Laevo was $25.2 \pm 4.3$ Nm and $30.6 \pm 9.5$ Nm respectively during the bending and lifting phases of the stoop tasks. Levio reduced the resistive torque during bending to $17.9 \pm 4.6$ Nm and offered increased assistance of $37.7 \pm 6.4$ Nm during lifting. Moreover, participants reported higher usability scores and lowest mean global discomfort score while using the Levio when compared to Laevo. These benefits demonstrates retrofitting passive exoskeletons is a viable direction for rapid prototyping of wearable exoskeletons for lifting.

Summaries are written automatically from each paper's text. Figures and videos come from the paper and its project page; all credit goes to the authors.