کنفرانس مهندسی برق ایران

صفحه اصلی / سی و چهارمین کنفرانس بین المللی مهندسی برق

A Full-Body IMU-Based Gait Dataset with Controlled Asymmetry Perturbations Featuring Acquisition Protocol and Baseline Validation

نویسندگان :

Hanieh Moradi¹ Yas Vaseghi² Arash Abbasi Larki³ Akram Shojaei⁴ Mehdi Delrobaei⁵

1- K. N. Toosi University of Technology 2- K. N. Toosi University of Technology 3- K. N. Toosi University of Technology 4- K. N. Toosi University of Technology 5- K. N. Toosi University of Technology

کلمات کلیدی :

Biomechatronic systems،Gait analysis،Kinematic data،Gait asymmetry،Neuromuscular coordination

چکیده :

We present a publicly available, full-body wireless inertial dataset capturing induced gait asymmetry in 11 healthy adults under controlled limb perturbations. Participants performed eight walk conditions, including normal, weighted dominant hand (1 kg), braced dominant leg (knee immobilizer), and combined, each at self-selected and fast speeds, yielding over 3,000 gait cycles. Data were recorded at 250 Hz using 10 synchronized IMUs placed on the trunk, pelvis, and upper and lower limbs. Joint kinematics were derived from segment orientations, and statistical analysis confirmed the efficacy of the perturbations. Wilcoxon Signed-Rank test results revealed significant reductions in knee flexion-extension range of motion during braced walking (p < 0.05) and altered shoulder ROM with added weight (p < 0.05). The dataset includes raw sensor outputs (accelerometer, gyroscope, and magnetometer), quaternions, Euler angles, and joint kinematics. This dataset provides a benchmark for developing signal processing, machine learning, and control algorithms for gait analysis and rehabilitation technologies.