Optimization-based Posture Prediction Research Articles

Optimal Posture and Supporting Hand Force Prediction for Common Automotive Assembly One-Handed Tasks

People often complete tasks using one hand for the task and one hand for support. These one-handed support tasks can be found in many different types of jobs, such as automotive assembly jobs. Optimization-based posture prediction has proven to be a valid tool in predicting the postures necessary to complete the tasks, but the related external support forces have been prescribed and not predicted. This paper presents a method in which the optimal posture and related supporting hand forces can be predicted simultaneously using optimization and stability analysis techniques. Postures are evaluated using a physics-based human performance measure (HPM) while external forces are assessed using stability analysis. The physics-based performance measures are based on joint torque. Stability is analyzed using criteria based on a 3D zero moment point (ZMP). The human model used in the prediction contains 56 degrees of freedom and is based on a 50th percentile female in stature. Tasks based on common automotive assembly one-handed tasks found in literature are considered as examples to test the proposed method. Overall, the predicted supporting hand forces have good correlation with experimentally measured forces.

Hybrid method for driver accommodation using optimization-based digital human models

Many applications exist where humans are required to perform a task in a seated position, such as operating a vehicle. Seated posture inside a vehicle influences driver performance and control of the vehicle. For people of extreme stature, tall or short, and for people of extreme width, obese or pregnant populations, it can be difficult to safely operate a vehicle if there is not enough room in the cab or if some controls cannot be reached. This study proposes a hybrid method for predicting the optimum driver seat adjustment range to accommodate diverse drivers in any vehicle based on a direct optimization-based posture prediction method. The proposed hybrid method combines three approaches: a boundary manikin approach, a population sampling approach, and a special population approach. The boundary manikin approach places two boundary manikins (5% female and 95% male) inside a virtual vehicle cab to perform tests. The population sampling approach spans a multitude of test subjects ranging in stature from 158 to 185 cm, determining the range from a plot of predicted hip point distances from the point of contact of the right heel and the floor. The special population approach studies the effect that size and shape changes, such as pregnancy, have on seated posture inside a vehicle. Also given is an indication of discomfort through the output values of the multi-objective function in the optimization formulation. A combination of the three approaches is used to determine an optimal adjustment range for the driver seat, thus allowing most people to safely operate the vehicle. Results of the simulations are validated using experimental determinations of the driver seat adjustment range from the literature. The main benefit of using this method is that the human aspect of design can be included early in the design process, thereby reducing or eliminating prototypes. Another benefit of the simulation is that it can be adapted to other seating tasks such as: occupant seat check inside a vehicle; workstation design; and issues related to other special populations such as obese individuals, dwarfs, and children.

Use of multi-objective optimization for digital human posture prediction

With sufficient fidelity, the use of virtual humans can save time, money, and lives through improved product design, process design, and understanding of behaviour. Optimization-based posture prediction is a unique tool, and this article presents a study that advances posture prediction with a multi-objective optimization (MOO) approach. MOO is used to both develop and combine the following human performance measures: joint displacement; musculoskeletal discomfort; and a variation on potential energy. The following MOO methods are studied in the context of human modelling: objective sum; min–max; and global criterion. Using MOO yields realistic results. Of the independent performance measures, discomfort generally provides the most accurate postures. Potential energy, however, is not a significant factor in governing human posture and should be combined with other performance measures. The three MOO methods for combining performance measures yield similar results, but the objective sum provides slightly more realistic postures.

Optimization-based posture prediction for human upper body

SUMMARYA general methodology and associated computational algorithm for predicting postures of the digital human upper body is presented. The basic plot for this effort is an optimization-based approach, where we believe that different human performance measures govern different tasks. The underlying problem is characterized by the calculation (or prediction) of the human performance measure in such a way as to accomplish a specified task. In this work, we have not limited the number of degrees of freedom associated with the model. Each task has been defined by a number of human performance measures that are mathematically represented by cost functions that evaluate to a real number. Cost functions are then optimized, i.e., minimized or maximized, subject to a number of constraints, including joint limits. The formulation is demonstrated and validated. We present this computational formulation as a broadly applicable algorithm for predicting postures using one or more human performance measures.

A New Methodology for Human Grasp Prediction

Grasping is an essential requirement for digital human models (DHMs). It is a complex process and thus a challenging problem for DHMs, involving a skeletal structure with many degrees-of-freedom (DOFs), cognition, and interaction between the human and objects in the environment. Furthermore, grasp planning involves not only finding the shape of the hand and the position and orientation of the wrist but also the posture of the upper body required for producing realistic grasping simulations. In this paper, a new methodology is developed for grasping prediction by combining a shape-matching method and an optimization-based posture prediction technique. We use shape matching to pick a hand shape from a database of stored grasps, then position the hand around the object. The posture prediction algorithm then calculates the optimal posture for the whole upper body necessary to execute the grasp. The proposed algorithm is tested on a variety of objects in a 3-D environment. The results are realistic and suggest that the new method is more suitable for grasp planning than conventional methods. This improved performance is particularly apparent when the nature of the grasped objects is not known <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">a</i> <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">priori</i> , and when a complex high-DOF hand model is necessary.