Abstract
The objective of the study was to evaluate the use
of a robot gripper in the assessment of mango (cv. BOsteen^)
firmness as well as to establish relationships between the nondestructive
robot gripper measurements with embedded accelerometers
in the fingers and the ripeness of mango fruit. Intact
mango fruit was handled and manipulated by the robot gripper,
and the major physicochemical properties related with
their ripening index were analyzed. Partial least square regression
models (PLS) were developed to explain these properties
according to the variables extracted from the accelerometer
signals. Correlation coefficients of 0.925, 0.892, 0.893, and
0.937 with a root-mean-square error of prediction of 2.524 N/
mm, 1.579 °Brix, 3.187, and 0.517, were obtained for the
prediction of fruit mechanical firmness, total soluble solids,
flesh luminosity, and ripening index, respectively. This research
showed that it is possible to assess mango firmness
and ripeness during handling with a robot gripper.