Predicting how humans would adapt their movement to previously unforeseen conditions is a grand challenge in biomechanics. Musculoskeletal simulations and optimization have the potential to meet this challenge and enable us to address a broad set of scientific and engineering questions. The goal of this dissertation was to develop simulations that predict human movement to study how muscular deficits contribute to gait pathologies and to aid in the design of assistive devices.

We developed simulation tools to study how muscle deficits contribute to gait pathologies. Deficits in the ankle plantarflexor muscles, such as weakness and contracture, occur commonly in conditions such as cerebral palsy, stroke, muscular dystrophy, and Charcot-Marie-Tooth disease. While these deficits likely contribute to observed gait pathologies, elucidating cause-effect relationships is difficult due to the often co-occurring biomechanical and neural deficits. To elucidate the effects of weakness and contracture, we systematically introduced isolated deficits into a musculoskeletal model and generated simulations of walking to predict gait adaptations due to these deficits. We developed a planar model containing nine degrees of freedom and eighteen musculotendon actuators, and an optimization framework through which we imposed simple objectives, such as minimizing cost of transport while avoiding falling and injury, and maintaining head stability. We first validated that our model could generate gaits that reproduced experimentally observed kinematic, kinetic, and metabolic trends for two cases:​ 1) walking at prescribed speeds between 0.50 m/s and 2.00 m/s and 2) walking at a self-selected speed (i.e., a speed determined by the optimization). We then applied mild, moderate, and severe levels of muscle weakness or contracture to either the soleus (SOL) or gastrocnemius (GAS) or both of these major plantarflexors (PF) and retrained the model to walk at a self-selected speed. The model was robust to all deficits, finding a stable gait in all cases. Severe PF weakness caused the model to adopt a slower, "heel-walking" gait, walking at a speed that was more than four standard deviations below the mean normal speed, and with a peak dorsiflexion during stance that was more than two standard deviations over the mean normal value. Severe contracture of only SOL or both PF yielded similar results;​ the model adopted a crouched, "toe-walking" gait, with minimum hip and knee flexion values that were more than two standard deviations over the mean normal value, and with peak dorsiflexion values in stance that were more than two standard deviations below the mean normal value. This work shows how simulations can elucidate fundamental mechanisms behind gait pathologies.

We also studied how simulations could be used to design assistive devices that enhance jumping performance. Technologies that augment human performance have recently become a focus of intensive research and development, driven by advances in wearable robotic systems. Success has been limited, however, due to the challenge of understanding human–robot interaction. To address this challenge, we developed an optimization framework to synthesize a realistic human standing long jump and used the framework to explore how simulated wearable robotic devices might enhance jump performance. A planar, five-segment, seven-degree-of-freedom model with physiological torque actuators was used to represent human musculoskeletal dynamics. The optimizer searched for physiological actuation patterns to maximize jump distance, yielding a simulated 2.27 m jump that captured salient kinematic and kinetic features of human jumps. We then tested an active augmentation design, modeled as a torque actuator that could apply a single pulse of up to 100 Nm of extension torque. This device was first added to a single joint at the ankle, knee, or hip, and the optimizer tuned the device to provide 100 Nm of torque over about a 200 ms duration, leading to jump distances that improved to between 2.49 m and 2.52 m. When the device was added to hip, knee, and ankle together, the optimizer tuned all three active devices to provide 100 Nm of torque over about a 200 ms duration, and the jump distance increased to 3.10 m. We next tested a passive device design by adding a rotational spring to the ankle, knee, and hip together. This design increased jump distance to 3.32 m by adding torques of up to 135 Nm, 365 Nm, and 297 Nm to the ankle, knee, and hip, respectively. This work highlights how simulation can aid in designing a device, allowing future researchers to iterate device designs more efficiently.

These studies showed how simulations that predict movement can be used to study cause-effect relationships in gait pathologies and inform device design. The methods presented here can serve as a guide for future research studies, and we provide our results, models, and simulation and optimization software as open-source resources to assist other researchers and accelerate their work.