Reproducibility and Variability of the Cost Functions Reconstructed From Experimental Recordings in Multifinger Prehension

ABSTRACT

The study examines whether the cost functions reconstructed from experimental recordings are reproducible over time. Participants repeated the trials on three days. By following Analytical Inverse Optimization procedures, the cost functions of finger forces were reconstructed for each day. The cost functions were found to be reproducible over time: application of a cost function C_i to the data of Day j (i≠j) resulted in smaller deviations from the experimental observations than using other commonly used cost functions. Other findings are: (a) the 2nd order coefficients of the cost function showed negative linear relations with finger force magnitudes; (b) the finger forces were distributed on a 2-dimensional plane in the 4-dimensional finger force space for all subjects and all testing sessions; (c) the data agreed well with the principle of superposition, i.e. the action of object prehension can be decoupled into the control of rotational equilibrium and slipping prevention.

Keywords:

• grasping
• inverse optimization
• prehension synergy
• principal component analysis
• reproducibility
• uniqueness theorem

ACKNOWLEDGMENTS

The authors thank Dr. A. V. Terekhov (Institut des Systèmes Intelligents et de Robotique, UPMC-CNRS, Paris, France) for useful comments on an early version of the manuscript, and Joel Martin for proofreading. They also greatly appreciate the helpful comments from the anonymous reviewers of this journal. The study was partially supported by National Institutes of Health grants AG-018751, NS-035032, and AR-048563.