Design and Validation of a Robotic Knee Exoskeleton with Musculoskeletal Simulation-Derived Biological Torque Control

¹Carnegie Mellon University ²Stanford University

Abstract

Data-driven controllers have enabled lower-limb robotic exoskeletons to adapt assistance to the user across many tasks. A leading example is biological torque control, which estimates the user's own joint moment from onboard sensors and assists in proportion to that estimate. However, its estimators are trained on data recorded with the target device, so deploying the controller on new hardware requires collecting new data. Existing open-source efforts release hardware, software, or controllers separately, so this data collection must still be repeated on the assembled device. Here, we present an open-source knee exoskeleton, built from off-the-shelf components and widely available fabrication tools, with a biological torque controller derived entirely from simulation. A muscle-driven imitation policy reproduces squatting motions from an open biomechanics dataset, and the resulting knee moments are distilled into an estimator that uses only onboard sensors, so no training data are recorded on the device. In seven able-bodied participants, assistance reduced peak biological knee extension moment by 21.5% and peak vastus medialis activation by 26.7% during squatting, relative to squatting without the exoskeleton (p<0.05). Our results demonstrate that a biological torque controller derived from simulation can transfer to physical hardware without device-specific data collection. Because the hardware and training pipeline are released together, research groups can train and compare knee exoskeleton controllers on shared hardware without additional data collection overhead.

Robotic Knee Exoskeleton Design

Figure 1. Overview of the robotic knee exoskeleton design. (a) Sagittal-plane view of the unilateral device, with straps, hooks, and cuffs at the thigh, shank, and ankle. (b) A participant wearing the bilateral device, shown left to right in sagittal standing, sagittal squatting, and frontal standing. (c) The high-level control board built around the embedded computer (top) and the low-level control board built around the microcontroller (bottom).

Table I. Device Specifications

Mechanical
Total mass, bilateral^* (kg)	3.5
Unilateral mass (kg)	1.3
Actuator mass (kg)	0.49
Rated / peak torque^** (Nm)	9 / 22
Range of motion (deg)	0–120
Compute/Electrical
Inertial sensor communication	SPI
Motor communication	CAN
Low-to-High level communication	CANFD
Low-level loop execution (µs)	~80
High/low-level round trip (ms)	<0.5
Estimated run time (h)	~2

^*Includes the battery and embedded electronics.
^**The stock motor driver limits output to 18 Nm per the manufacturer's safety constraint.

Simulation-Derived Biological Torque Control

Figure 2. Overview of the simulation-derived biological torque controller. An imitation policy is trained with reinforcement learning in a musculoskeletal simulation to reproduce reference squatting motion, using an imitation reward as its learning signal. Knee muscle moments from the policy rollouts are then distilled into a temporal convolutional network that estimates the knee moment from onboard sensor signals, trained by minimizing the error against the simulated moments. On the exoskeleton, the estimated knee moment is scaled and delayed to generate assistive torque.

Experimental Validation

Figure 3. Experimental setup for validating the integrated exoskeleton and controller during squatting. A participant wearing the exoskeleton on the split-belt instrumented treadmill, whose embedded force plates measure ground reaction forces, with example placements of the electromyography sensors and reflective motion capture markers.

Results

Figure 4. Group-averaged knee kinematics, kinetics, and muscle activation during squatting. (a) Knee angle, (b) biological knee extension moment with commanded torque, (c) vastus medialis activation, and (d) medial gastrocnemius activation over the normalized squat cycle. Solid curves represent group means across participants, and shaded regions represent ±1 standard deviation. Blue and brown indicate squatting without (No Exo) and with (Exo) the exoskeleton, respectively, and the dashed gray curve indicates the commanded torque with the exoskeleton. Muscle activation is expressed as a percentage of each participant's maximum voluntary contraction (%MVC).

Figure 5. Peak biological knee extension moment and muscle activation with and without the exoskeleton. (a) Peak biological knee extension moment, (b) peak vastus medialis activation, and (c) peak medial gastrocnemius activation without (No Exo) and with (Exo) the exoskeleton. Bars represent group means, and error bars represent ±1 standard deviation across participants. Asterisks indicate a significant difference between conditions (p<0.05, paired t-test).

Table II. Comparison with Existing Knee Exoskeleton Platforms

Device	Validation Subjects	Device Mass (kg)	Peak Torque (Nm)	Tasks	Reported Reduction
This Study	7 able-bodied	3.5	18	Squatting	26.7% vastus medialis activation
Nesler et al.	1 able-bodied	1.6^a	30	Lifting and lowering	78.9% vastus medialis activation
Lee et al.	10 able-bodied	4.4	15	Level walking, ramps, and stairs	6.2% metabolic cost^c
Zhang et al.	6 able-bodied	3.2	18	Level walking, ramp and stair ascent	4.3–12.0% metabolic cost
Boon et al.	8 able-bodied	2.6^a	55	Stair ascent	26% vastus medialis activation
Chen et al.	2 able-bodied	4.1	35	Kneeling	18.4% vastus medialis activation^b,c
Huang et al.	8 able-bodied	3.5	14	Level walking	19.6% vastus medialis activation^c
Sarkisian et al.	1 post-stroke	3.6^a	40	Sit-to-stand	29.3% vastus medialis activation
Yu et al.	7 able-bodied	3.5	54	Squatting	38.1% vastus medialis activation, 12.8% metabolic cost

Mass is for a bilateral device unless noted. Activation reductions are in peak values and relative to no exoskeleton unless noted. ^aUnilateral device. ^bMean rather than peak activation. ^cRelative to the unpowered or zero-torque exoskeleton.

@article{joshi2026KneeExo, title = {Design and Validation of a Robotic Knee Exoskeleton with Musculoskeletal Simulation-Derived Biological Torque Control}, author = {Joshi, Rajiv V. and An, Jimin and Park, Ilseung and Lee, Dawit and Kang, Inseung}, journal = {arXiv preprint arXiv:2610.41981}, year = {2026} }