Human-Arm-and-Hand-Dynamic Model With Variability Analyses for a Stylus-Based Haptic Interface

Abstract

Haptic interface research benefits from accurate human arm models for control and system design. The literature contains many human arm dynamic models but lacks detailed variability analyses. Without accurate measurements, variability is modeled in a very conservative manner, leading to less than optimal controller and system designs. This paper not only presents models for human arm dynamics but also develops inter- and intrasubject variability models for a stylus-based haptic device. Data from 15 human subjects (nine male, six female, ages 20-32) were collected using a Phantom Premium 1.5a haptic device for system identification. In this paper, grip-force-dependent models were identified for 1-3-N grip forces in the three spatial axes. Also, variability due to human subjects and grip-force variation were modeled as both structured and unstructured uncertainties. For both forms of variability, the maximum variation, 95 %, and 67 % confidence interval limits were examined. All models were in the frequency domain with force as input and position as output. The identified models enable precise controllers targeted to a subset of possible human operator dynamics.

End-point impedance measurements at human hand during interactive manual welding with robot

“…Identifying the mass, damping, and stiffness parameters together in a consistent way is mostly performed by using long duration data, such as 300 ms [20], 370 ms [24], 30 sec [10,13], and 1 minute [15]. In our work we aim at estimating the impedance values using a time window of 200 ms.…”

“…The tacit skills are learned implicitly first in such practice hours and then throughout the initial years of professional work [3]. The impedance measurement techniques developed in domains such as human physiology, physiological rehabilitation, and human-robot interaction [8][9][10][11][12][13][14][15] might be applied for quantifying the tacit skills of welding. This study is a first attempt to perform impedance measurements with professional and novice welders.…”

“…The present study focuses on this force interaction by analyzing how human hand reacts to disturbances. Human arm joint impedance or human end-point hand impedance can be measured by introducing either position [8,9,10,11] or force [12,13,14,15] disturbances. In this work we follow the force disturbance approach; because it is conveniently applied for manual welding, where the welding line is not pre-programmed and it might even dynamically change according to the working style of the welder.…”

“…holding a handle at a reference position [10,13,15,16,17,18], or trajectory control, while the hand is moving a handle from one point to another [8,9,12]. In the latter case, a reference trajectory is used either for calculation of a reference position [8,9] or for guiding the movement with a visual interface [12].…”

“…Through learning, the subjects increase their end-point stiffness in order to counter the divergent force field. In [15], the authors measure the endpoint impedance while the subjects apply different grasp force on an object while positioning it at a given location.…”