UMich 2021 Dataset¶

The Effect of Walking Incline and Speed on Human Leg Kinematics, Kinetics, and EMG¶

Overview¶

Brief Description: Comprehensive treadmill-based locomotion dataset including walking at various speeds and inclines, with synchronized kinematics, kinetics, and EMG data. This dataset focuses on continuous parameterization of human gait across different walking conditions.

Collection Year: 2018-2021
Dataset Size: ~1.2 GB (parquet format)
License: Open Access (IEEE DataPort)

Institution: University of Michigan, Department of Robotics, Mechanical Engineering, and Electrical and Computer Engineering

Principal Investigators: Robert D. Gregg IV, Ph.D. (Locomotor Control Systems Laboratory)

Citation Information¶

Primary Citation¶

@misc{embry2018walking,
  title={The effect of walking incline and speed on human leg kinematics, kinetics, and EMG},
  author={Embry, K. and Villarreal, D. and Macaluso, R. and Gregg, R.D.},
  year={2018},
  publisher={IEEE DataPort},
  doi={10.21227/gk32-e868}
}

Associated Publications¶

Embry, K., Villarreal, D., Macaluso, R., & Gregg, R.D. (2018). "The Effect of Walking Incline and Speed on Human Leg Kinematics, Kinetics, and EMG"
IEEE DataPort. DOI: 10.21227/gk32-e868
Macaluso, R., Embry, K., Villarreal, D., & Gregg, R.D. (2020). "Human Leg Kinematics, Kinetics, and EMG During Phase-Shifting Perturbations at Varying Inclines"
IEEE DataPort. DOI: 10.21227/12hp-e249
Elery, T., Rezazadeh, S., Reznick, E., Gray, L., & Gregg, R.D. (2020). "Effects of a Powered Knee-Ankle Prosthesis on Amputee Hip Compensations: A Case Series"
IEEE DataPort. DOI: 10.21227/sngq-4x29
Best, T.K., Embry, K.R., Rouse, E.J., & Gregg, R.D. (2023). "Phase-Variable Control of a Powered Knee-Ankle Prosthesis over Continuously Varying Speeds and Inclines"
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

Acknowledgments¶

This research was supported by: - NIH Director's New Innovator Award (2013) - $2.3 million over 5 years for phase-based control research - NIH R01 Grant (2018) - $2.2 million for investigation of agile powered prosthetic leg control - NIH R01 Grant (2021) - $1.7 million for design and control of modular powered orthoses

Grant Details: - NIH RePORTER - DP2HD080349 - NIH RePORTER - R01HD094772

Dataset Contents¶

Subjects¶

Total Subjects: 10 (UM21_AB01 - UM21_AB10)
Subject ID Format: UM21_AB## (Dataset: UMich 2021, Population: Able-bodied)
Demographics:
Age Range: 20-60 years
Sex Distribution: 5F/5M
Height Range: 1617-1900 mm
Weight Range: 53.7-87.0 kg
Mean Age: 30.4 years
Mean Weight: 74.63 kg
Mean Height: 1727.8 mm
Population: All able-bodied (AB) healthy adults

Tasks Included¶

Task ID	Task Description	Duration/Cycles	Conditions	Notes
level_walking	Level walking	60 sec trials	0° at 0.8, 1.0, 1.2 m/s	Bertec treadmill
incline_walking	Incline walking	60 sec trials	+5°, +10° at multiple speeds	Bertec treadmill
decline_walking	Decline walking	60 sec trials	-5°, -10° at multiple speeds	Bertec treadmill
running	Running (if included)	60 sec trials	1.8-2.4 m/s	Level only
transitions	Speed/incline transitions	Variable	Continuous changes	Phase-shifting

Data Columns (Standardized Format)¶

Variables: 45+ biomechanical features
Kinematics: Hip, knee, ankle angles (sagittal, frontal, transverse)
Kinetics: Joint moments and powers
EMG: 7 channels (gluteus medius, rectus femoris, vastus lateralis, biceps femoris, tibialis anterior, gastrocnemius medialis, soleus)
Ground reaction forces: Vertical and horizontal components
Format: Phase-indexed (150 points per gait cycle)
File: converted_datasets/umich_2021_phase.parquet
Units:
Angles: radians
Moments: Nm/kg (normalized)
Powers: W/kg (normalized)
EMG: Normalized to maximum voluntary contraction
Forces: N/kg (normalized)

Data Collection Methods¶

Equipment Specifications¶

Treadmill: Bertec split-belt instrumented treadmill
Force plates: Dual 6-DOF force platforms (1000 Hz)
Speed range: 0-3.0 m/s
Incline range: -15° to +15°
Motion Capture: Vicon system with 12 cameras (200 Hz)
EMG System: Delsys Trigno wireless (2000 Hz)
Marker Set: Modified Helen Hayes with additional tracking markers

Processing Pipeline¶

Inverse Kinematics: Visual3D software
Filtering: 4th order zero-lag Butterworth
Kinematics: 6 Hz cutoff
Kinetics: 25 Hz cutoff
EMG: 20-450 Hz bandpass, RMS envelope
Cycle Detection: Heel strike from vertical GRF
Phase Variable: Monotonic progression through gait cycle
Normalization: Time-normalized to 150 points using cubic spline

Research Focus¶

Phase-Variable Control¶

The Locomotor Control Systems Lab pioneered phase-variable control for powered prostheses, using a continuous phase variable that robustly represents gait cycle timing across different walking conditions. This approach enables smooth transitions between different speeds and inclines without discrete mode switching.

Clinical Translation¶

The lab's research has led to successful clinical trials of powered knee-ankle prostheses, demonstrating improved metabolic efficiency and biomechanical symmetry compared to passive devices.

Contact Information¶

Dataset Curator: Robert D. Gregg IV, Ph.D.
Lab Website: https://locolab.robotics.umich.edu/
Lab Email: locolab@umich.edu
Technical Support: Contact via lab email
IEEE DataPort: Browse Datasets

Funding Acknowledgment¶

This dataset was collected with support from: - NIH Director's New Innovator Award (DP2HD080349) - NIH R01 Grant for agile powered prosthetic legs (R01HD094772) - Burroughs Wellcome Fund Career Award at the Scientific Interface ($500,000)

Additional Resources: - University of Michigan Robotics - Michigan Engineering News

Lab Description¶

The Locomotor Control Systems Laboratory is a highly interdisciplinary environment dedicated to scientific innovation, clinical translation, and individual career development. The lab develops high-performance control systems for robotic prostheses and orthoses to enable mobility and improve quality of life for persons with disabilities. The research approaches this needs-driven work from the perspective of dynamical systems and control theory, investigating and translating concepts from legged robotics into transformative solutions for physical rehabilitation.

Dr. Gregg's research interests include actuator design and control, nonlinear control theory, and biomechanics of human locomotion. He is most known for his works on bipedal locomotion control and wearable robotics, with research focusing on integrating human biomechanics insights into robot control algorithms to improve natural movement efficiency.

Lab Publications: Google Scholar | ResearchGate

Usage¶

from user_libs.python.locomotion_data import LocomotionData

# Load the dataset
data = LocomotionData('converted_datasets/umich_2021_phase.parquet')

# Get data for analysis
cycles_3d, features = data.get_cycles('SUB01', 'level_walking')

Data Validation¶

📊 Validation Status¶

Validation Configuration: - Ranges File: default_ranges.yaml - SHA256: 76ab6a11... (first 8 chars) - Archived Copy: umich_2021_phase_2025-08-07_221225_ranges.yaml

Metric	Value	Status
Overall Status	99.8% Valid	✅ PASSED
Phase Structure	150 points/cycle	✅ Valid
Tasks Validated	3 tasks	✅ Complete
Total Checks	396,096	-
Violations	979	⚠️ Minor