Human Motion Data in Complex Construction Environments Setup with and without Knee Exoskeleton Assistance

Construction work involves complex, non-repetitive tasks that increase risks of work-related musculoskeletal disorders (WMSDs) and productivity loss. This study presents an integrated evaluation framework combining wearable knee exoskeleton assistance, immersive virtual reality (VR) task scenarios, and synchronized multi-sensor measurements to quantify safety- and performance-related outcomes in construction-like settings. Using a modular stair–ramp–deck platform, twenty-one adults completed multi-task protocols with and without knee exoskeleton assistance in a randomized crossover design while the motion capture system, force plates, electromyography (EMG), inertial measurement units (IMUs), and metabolic sensing recorded biomechanics, stability, and effort. Assistance reduced lower-limb muscular demand and improved postural stability during kneeling and VR precision tasks. User surveys indicated perceived stability gains but highlighted comfort and weight concerns. The framework enables task-relevant, reproducible assessment of safety and productivity in realistic construction environments. All multimodal data and protocols are released to support benchmarking and future task-aware assistance systems.