Versatile Exoskeleton Control Supports Elderly Joint Energetics During Hip-Intensive Tasks

Jiefu Zhang; Nikhil V. Divekar; Chandramouli Krishnan; Robert D. Gregg

cs.ROMar 23, 2026

Versatile Exoskeleton Control Supports Elderly Joint Energetics During Hip-Intensive Tasks

Authors: Jiefu Zhang, Nikhil V. Divekar, Chandramouli Krishnan, Robert D. Gregg

Organizations: Robotics Department, University of Michigan, Ann Arbor, MI 48109, USA

Abstract

Age-related mobility decline is frequently accompanied by a redistribution of joint kinetics, where older adults compensate for reduced ankle function by increasing demand on the hip. Paradoxically, this compensatory shift typically coincides with age-related reductions in maximal hip power. Although robotic exoskeletons can provide immediate energetic benefits, conventional control strategies have limited previous studies in this population to specific tasks such as steady-state walking, which do not fully reflect mobility demands in the home and community. Here, we implement a task-agnostic hip exoskeleton controller that is inherently sensitive to joint power and validate its efficacy in eight older adults. Across a battery of hip-intensive activities that included level walking, ramp ascent, stair climbing, and sit-to-stand transitions, the exoskeleton matched biological power profiles with high accuracy (mean cosine similarity 0.89). Assistance significantly reduced sagittal plane biological positive work by 24.7% at the hip and by 9.3% for the lower limb, while simultaneously augmenting peak total (biological + exoskeleton) hip power and reducing peak biological hip power and torque. These results suggest that hip exoskeletons can potentially enhance endurance through biological work reduction, and increase functional reserve through total power augmentation, serving as a promising biomechanical intervention to support elderly mobility.

Figures & tables

Fig. 1: Overview of the physics-informed task-agnostic exoskeleton controller structure. Sensor measurements are used to generate velocity-modulated virtual spring torque bases, which are subsequently modulated based on task context to generate the output torque.

Panel of Figure 2 — Fig. 2: Left: Definitions of angles used in the controller. Black-outlined circles represent hip and knee joint centers. $\theta_{\text{torso}},\theta_{\text{thigh,l}},\theta_{\text{thigh,r}},\theta_{\text{ips}}$ , and $\theta_{\text{con}}$ denote the angles of the torso, left and right thigh, ipsilateral hip, and contralateral hip, respectively. $\theta_{\text{diff}}$ represents the inter-thigh angle difference. Hip flexion is defined as the positive direction. Figure reproduced from [ 41 ] with permission. Right: Picture of the modified M-BLUE hip module: (1) Battery, (2) Inertial measurement unit (IMU), (3) Thigh interface, (4) Electronics Box, (5) Post-op Brace, (6) Actuator. For clarity only one of the IMUs, actuators, and thigh interfaces are labeled.

Activity	LG 0.85	LG 1.15	RA 5.2	RA 11	SA 5	SA 7	RD 5.2	RD 11	SD 5	SD 7
SIM	0.9751	0.9805	0.9815	0.9802	0.9661	0.9672	0.7029	0.7785	0.7567	0.6843

TABLE I: Cosine similarity (SIM) between exoskeleton power and normative biological hip power (SIM = 1 indicates perfect similarity).

Fig. 3: Normalized biological hip moment and exoskeleton assistance torque over a cycle of activities. No EXO (Bio.) stands for the biological hip moment in the No EXO condition, EXO (Bio.) stands for the biological hip moment in the EXO condition, and EXO (Assist.) stands for the commanded torque of the exoskeleton in the EXO condition. Positive values represent flexion moment/torque. Each line represents inter-participant average, and each shaded region represents

\pm 1

standard deviation.

Fig. 4: Extension torque scale across activities. Markers represent the inter-participant average scale, and error bars represent

\pm 1

standard deviation. Note that for SA and SD, the inclination angles are approximated from the horizontal and vertical changes from the bottom to the top of the staircase.

Fig. 5: Per-limb, mass-normalized peak biological hip moment across hip-intensive activities under No EXO and EXO conditions. Boxplots show the inter-participant median (horizontal line), interquartile range (box), inter-participant mean (black square), and whiskers to the most extreme observation within 1.5 times of the interquartile range. Numbers indicate the percent reduction in EXO relative to No EXO. Asterisks indicate statistical significance (Holm–Bonferroni-corrected

p\leq 0.05

Fig. 6: Normalized biological hip power and exoskeleton assistance power over a cycle of activities. No EXO (Bio.) and EXO (Bio.) represent the biological hip power in the No EXO and EXO conditions, respectively, and EXO (Assist.) represents the exoskeleton assistance power in the EXO condition. Lines and shaded regions represent the inter-participant average and

\pm 1

standard deviation, as in Fig. 3 .

Activity	LG 1.0	LG 1.2	RA 5	RA 10	SA	STS	RD 5	RD 10	SD
SIM	$0.86\pm 0.06$	$0.83\pm 0.05$	$0.88\pm 0.07$	$0.94\pm 0.05$	$0.91\pm 0.04$	$0.91\pm 0.08$	$0.80\pm 0.10$	$0.73\pm 0.16$	$-0.11\pm 0.43$

TABLE II: Cosine similarity (SIM; average

\pm

standard deviation) between exoskeleton and biological hip power during the EXO condition. SIM = 1 indicates perfect similarity.

Fig. 7: Per-limb, normalized peak biological hip positive power and total (biological + exoskeleton) hip positive power in No EXO and EXO conditions. Box-and-whisker conventions are as in Fig. 8 . Numbers denote percent change in EXO relative to No EXO. Asterisks indicate statistical significance (Holm–Bonferroni-corrected

p\leq 0.05

Fig. 8: Per-limb, mass-normalized biological positive work at the hip and for the lower limb (hip + knee + ankle) across hip-intensive activities under No EXO and EXO conditions. Boxplots show the inter-participant median (horizontal line), interquartile range (box), inter-participant mean (black square), and whiskers to the most extreme observation within 1.5 times of the interquartile range. Numbers indicate the percent reduction in EXO relative to No EXO. Asterisks denote significant EXO vs No EXO differences within an activity (Holm–Bonferroni corrected

p\leq 0.05

). Task abbreviations (used throughout Figs. 8 – 6 ): LG, level walking; RA5/RA10, ramp ascent at 5 ∘ /10 ∘ ; SA, stair ascent; STS, sit-to-stand.

Hip Exoskeleton Control

See all 10 papers

Effects of Assistance Delay on Joint Mechanics and Energetics in Biological Torque Control of a Hip Exoskeleton
Sep 21, 2026Jimin An, Ryan Lee, Jingshu Peng +2Robotic Control Hip Exoskeleton Control
Department of Mechanical Engineering, Carnegie Mellon University, Pittsburgh, PA, 15213 USA
Staged Multi-Agent Training (SMAT) for Hip Exoskeletons: Metabolic and Biomechanical Validation of a Simulation-Trained Co-Adaptive Controller
Aug 1, 2026Yifei Yuan, Jakob Wolf, Ghaith Androwis +1Robotic Control Hip Exoskeleton Control
Department of Biomedical Engineering, New Jersey Institute of Technology, Newark, NJ 07102, USA · Kessler Foundation, West Orange, NJ 07052, USA
Learning Hip Exoskeleton Control Policy via Predictive Neuromusculoskeletal Simulation
Mar 4, 2026Ilseung Park, Changseob Song, Inseung KangHip Exoskeleton Control Robotic Control
Department of Mechanical Engineering, Carnegie Mellon University, Pittsburgh, PA, 15213 USA.

Sep 21, 2026J/K move · Enter open · S save

Effects of Assistance Delay on Joint Mechanics and Energetics in Biological Torque Control of a Hip Exoskeleton

Jimin An, Ryan Lee, Jingshu Peng, Eni Halilaj+1

Department of Mechanical Engineering, Carnegie Mellon University, Pittsburgh, PA, 15213 USA

Robotic Control Hip Exoskeleton Control Assistive Robotics

Open paper AskPDF arXiv

Biological torque control directly maps an estimated human joint moment to exoskeleton assistance, providing a task-agnostic strategy for supporting diverse locomotor activities. However, it remains unclear whether a fixed state-to-torque mapping provides effective assistance across biomechanically distinct tasks. We examined how assistance delay affected hip exoskeleton performance during level-ground (LG), ramp-ascent (RA), and ramp-descent (RD) walking. Eight participants completed a zero-torque baseline condition and five active assistance conditions with delays ranging from 40 to 320 ms. Across tasks and active delays, assistance reduced net metabolic rate by 5.24%, positive biological hip joint work by 5.86%, and total lower-limb positive joint work by 1.68% (all p < 0.05). Assistance delay affected both joint-work outcomes (both p < 0.001) but not net metabolic rate. Mechanical unloading generally decreased with increasing delay, whereas metabolic benefits remained comparatively stable. Relative to the zero-torque condition, net metabolic rate decreased by 9.75% during LG and 7.20% during RA but increased by 1.23% during RD. We did not detect task-dependent differences in the delay response. Our findings indicate that biological torque mappings should be evaluated based on the target outcome and mechanical role of the assisted joint, and that predominantly positive-power assistance may not generalize to negative-work-dominant locomotion without modification.

Abstract

Hip Exoskeleton Control

Explore similar work

Effects of Assistance Delay on Joint Mechanics and Energetics in Biological Torque Control of a Hip Exoskeleton

Staged Multi-Agent Training (SMAT) for Hip Exoskeletons: Metabolic and Biomechanical Validation of a Simulation-Trained Co-Adaptive Controller

Learning Hip Exoskeleton Control Policy via Predictive Neuromusculoskeletal Simulation