MODEL OF A HUMAN WRIST FOR ESTIMATING FORCE FEEDBACK IN HUMAN-ROBOT INTERACTIONS

This research is part of the musculoskeletal tele-robotic imaging machine (MSK-TIM) project. The MSK-TIM is a device that facilitates remote ultrasound diagnosis through remote control of an ultrasound probe. The work reported here is to facilitate human- robot interaction in remote ultrasound imaging. To reduce the overall examination time, areas of improvement include improving the intuitiveness of control via haptic feedback, mitigating delay, and improving training. This work investigates developing a virtual model which can be used for these purposes. Developing a model to understand the biomechanics of the patient's wrist is important in several fields including medicine and robotics. Up to now, most models have been developed to accurately represent the complex internal structure of the wrist. However, there are several applications, such as generating a force for haptic feedback, informing a predictive controller to mitigate delay, or providing a simulated training environment, which requires a rapid solution. This study reports on investigation on how we developed a finite element wrist model to test the feasibility of a force-feedback control for a remote ultrasound system. The geometry and material parameters of the human wrist were obtained in vivo. The finite element model was then incrementally modified to improve the computational time while measuring the corresponding error. As a result, the computation time was reduced by 95%. As a result of this study, it has been shown that soft tissues can be generalised as a phenomenological material to decrease model complexity. The average first-order Ogden parameters for soft tissue in the wrist were derived to be (mu, alpha 1) = (4.59 kPa, 10.69). Finally, the geometry of anatomy can be simplified, without a major reduction in accuracy, to greatly reduce the computation time.