Digital human posture and motion prediction considering cognitive decision making

Every day, humans are presented with tasks that they complete with little effort of even consideration of the planning that goes into the movement. Movements such as reaching around an obstacle or manual manipulation tasks are completed with ease, even though the complexities and years of learned behavior are largely hidden from the person. Digital Human Modeling (DHM) and specifically optimization-based posture and motion prediction methodologies have employed numerical methods in order to simulate/predict/analyze human movements. However, these movements are heavily constrained such that the panning of the motion/posture is explicitly provided in the formulation of the problem. This implies that for each posture or movement under analysis, a unique formulation that relies heavily on the experience of the researcher to provide these constraints is required. However, there has been much study within the realm of cognitive psychology focused on the reasoning or motivation behind the cognitive planning of movements. This presents an opportunity for DHM to adopt these methodologies in order to provide a more general or versatile posture and motion prediction framework. This work presents the addition of cognitive principles into the optimization-based posture and motion prediction formulations. It considers two specific scenarios. First, the simulation/prediction of manual manipulation tasks are considered such that a single formulation can accomplish multiple tasks. It adopts a theory from cognitive psychology referred to as the end-state comfort effect in order to derive general constraints for the prediction of the initial and final posture states that frame the movement related to the manual manipulation task. It considers multiple tasks from the literature that have been heavily studied through experimentation in order to evaluate the efficacy of the formulation. The results show strong correlation with observations reported in the literature. Second, the formulation of a collision avoidance algorithm considering perceived risk is offered. Humans tend to overestimate the risk associated with colliding with objects during movement, and therefore adopt suboptimal movements with respect to biomechanical cost. An experiment in order to evaluate human performance when avoiding obstacles in movement is designed. Using Bayesian inference, perceived risk, which is studied heavily in motor planning research, is modeled and minimized for use as a constraint in the larger human motion prediction problem. The performance using the new formulation is compared to the observed performance from the experiment. The results show that the new formulation can account for the suboptimal behavior observed in real subjects while still optimizing biomechanical cost.