Agriculture Reference
In-Depth Information
(a)
(b)
FIGURE 7.31
(a) 7 DOF articulated redundant manipulator configuration, workspace and
dexterity analysis model, and (b) harvesting trajectories model. (From Sivaraman, B., Design
and development of a robot manipulator for citrus harvesting, PhD dissertation, Univ. of
Florida, 2006. With permission.)
singularity analysis that was done in MATLAB robotics toolbox that enabled link
parameter selection based on workspace requirements. Figure 7.31c depicts the tool
space dexterity analysis completed in RobotectPro where the dexterity ellipse dem-
onstrates excellent tool dexterity. Finally, Figure 7.31d shows a harvesting workspace
scenario where various harvesting trajectories were generated for each crossing point
on the canopy map. These were used to determine the required actuator torques,
velocities, and accelerations necessary to achieve desired harvesting cycle rates.
7.4.3 R
OBOTIC
S
YSTEMS
D
EVELOPMENT
In 2001, the University of Florida and the Florida Department of Citrus began a
collaborative investigation into the potential of using robotics for citrus harvest-
ing. Through a synthesis of prior effort successes and failures several observations
were made: (1) fruit detection rates are inadequate to achieve commercial feasibility;
(2)Â harvesting efficiency as it relates to both fruit detection and fruit removal is too
low; (3) fruit located within the canopy interior is very difficult to harvest; and (4)
current end effector technologies are inadequate for the general harvesting problem,
especially in fruit where the stem must be clipped. In the past three decades, there
have been numerous technological advances that have improved the potential for
agricultural robotics. The cost and speed of computers have vastly improved, redun-
dant manipulators have been developed, the cost and performance of solid-state
sensors such as the color CCD camera became available in the market, computer
algorithms have improved to match computing speeds, and numerous other advances
have been made in actuator design, mobile power, hyper-redundant manipulators,
prosthetics, and so forth—all of which provides encouragement that agricultural
robotic harvesting may be on the horizon.
