Worker fatigue increases the risk for illnesses and injuries. An estimated annual cost in the US of over $130 billion is from fatigue-related lost productive work time to employers, which indicates that fatigue needs to be considered during the workplace safety design process. Although different muscle fatigue models have been developed, all were applied to isometric contractions (contractions without muscle shortening), but the majority of everyday activities are involved in concentric (muscle contracts/shortens) and eccentric (muscle lengthens/returns to resting state) muscle movements, i.e., repetitive dynamic tasks. Conventional motion simulation approaches for injury prevention typically optimize a motion without considering muscle fatigue. Thus, the goal of this project is to address the need for a musculoskeletal model that can predict muscle movement considering muscle fatigue. The model can be adapted to the physical properties of an individual worker, e.g., height, weight, length of body segments, etc. The methods and associated numerical tools developed will be applicable to broad occupational health and safety designs such as lower back injury prevention for repetitive lifting and repetitive package handling in the delivery industry. The project will also enable education and training for undergraduates, graduate students, and store employees. Results will be integrated into courses for Biomechanics and Digital Human Modeling and made available for future generations of engineers. In addition, a week-long summer camp will be organized for local store managers with lifting jobs at Texas Tech University. For these managers this summer camp will help them to understand the injury mechanism and causes for injuries, contributing to their awareness of work-related injuries whenever employees conduct repetitive lifting tasks daily.
The goal of this project is to develop a novel and efficient dynamic motion prediction tool considering muscle fatigue for repetitive dynamic tasks, which, if successful, will be the first full body biomechanics human model with this capability. The project’s objectives are to: develop: a new joint space muscle fatigue model for repetitive dynamic tasks; develop a new joint space predictive simulation method considering fatigue; and decompose the fatigued joint torques into fatigued muscle forces. The Research Plan is organized under 6 tasks. TASK 1 is to develop a three-compartment joint space fatigue model for repetitive tasks beginning with a 3D musculoskeletal model that has 30 DOFS, 21 segments, 324 musculotendon actuators and 5 lumbar vertebrae connected with 6 DOF joints. Fatigue is incorporated by bundling all muscles responsible for each joint into one virtual muscle with virtual units being divided into compartments depending on the state in which they are in: active, fatigued or resting. TASK 2 is to perform an inverse dynamics-based optimization without fatigue. The design variables for the skeleton optimization problem are joint angles from which joint torques can be computed. TASK 3 is to optimize join space motion prediction considering fatigue using collocation methods. The joint torques obtained under Task 2 will be used to calculate a “target load†vector that will be used to initiate the fatigue process. The optimization problem is to find the optimal joint angles, joint torques, joint resultant and virtual muscle active states that minimize the cost function of normalized joint torque squared subject to model dynamics equations of motion. TASK 4 is to find the lower extremity and lumbar spine model muscle forces corresponding to the fatigued joint torques using static optimization. Tasks 1-4 are interconnected elements of a complete nonlinear motion optimization considering muscle fatigue for dynamic tasks. TASK 5 is to collect experiment related data from 20 subjects (10 males and 10 females) of varying ages, statures, and BMIs. The data will be used to validate the joint space muscle fatigue model and skeletal motion prediction. TASK 6 is to validate the muscle fatigue model for repetitive dynamic tasks involving wrist, elbow, shoulder, trunk hip, knee and ankle joints and then to validate the 3D motion prediction model considering muscle fatigue during a repetitive box lifting process. For joint related validations, 8 subjects of each gender will be used to tune the model and the 2 remaining subjects will be used for validation. For 3D motion prediction validation, three aspects (muscle levels, joint profiles and ground reaction forces), model predictions will be compared to aspects determined from EMG and motion capture data.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
