An approach to sensor fusion in medical robots

Abstract

This paper proposes a multi-sensor data fusion technique to determine the complex interactions between the sensory, muscular and mechanical components of the human locomotor systems (neuromechanics). We investigated the use of an array of accelerometers, rate gyroscopes, force plates and electromyogram for the assessment of x-y-z components of the hip, thigh, shank and foot acceleration and velocity in the sagittal plane. The objective here is to determine the impact factor each body segment's acceleration has in terms of producing a functional gait. Six able-bodied subjects were employed in this experiment; and subject walked at three different speeds and his/her dynamic/kinematic behaviors of the lower limb recorded with the sensors afore mentioned. Eight distinct gait variability for each subject recorded was detected by our sensing system. Fuzzy fusion technique was employed to evaluate the empirical results and the output matrix shows the relation between the body segments in question in terms of their x-y-z acceleration components. The implementation of this fusion matrix will enhance modeling and building medical robots intended for paraplegic rehabilitation as well as intelligent mobile robots for effective industrial manufacturing.