Agriculture Reference
In-Depth Information
estimation. Correct estimation of the range allows the robot to properly position the
fruit inside the end effector. This could be compensated for by adding a fruit proxim-
ity sensor in the grip of the end effector to provide positive feedback that the fruit
is in harvesting position before the gripper closes. Several factors contributed to the
erroneous range estimation. First, the ultrasonic sensor's analog range signal is rather
noisy due to irregularity in the surface of the tree and the fruit. Although an attempt
was made to compensate for this using stereo vision, difficulty in corresponding the
features in the image pair can also cause erroneous reading. Other causes of the
failed attempts were the inability to grab the fruit due to leaf occlusion and cluster-
ing, which is partially at least attributed to the end effector design. These results
have demonstrated the feasibility of robotic fruit harvesting, while also illustrating
the challenges that have hindered successful development and commercial adoption.
7.5 AUTONOMOUS VEHICLE GUIDANCE
Autonomous navigation is a valuable tool in agricultural systems development
because it has the potential to significantly improve steering accuracy and repeat-
ability, while freeing the operator to attend to higher level activities. In orchard and
grove applications, autonomous guidance has several potential uses ranging from
mowing, spraying, disease and pest scouting, planting, and harvesting. Most efforts
to date have focused on semiautonomous operations where human supervisors would
be onboard the vehicle during operation. However, future opportunities exist for
fully autonomous systems, although personnel safety, cost, reliability, and legal con-
straints may delay the adoption of fully autonomous operations in groves.
Vehicle position, heading, steering effort, and speed with respect to the desired path
are the most important issues that must be considered. GPS in combination with inertial
navigation systems has been widely used as positioning and heading sensors in tradi-
tional field agriculture application. Both RTK GPS and real-time DGPS have been tested
with success based on the degree of accuracy required in the navigation system. There
is a trade-off between accuracy and cost in the selection of DGPS and RTKGPS, with
the latter being more accurate and expensive. RTK GPS has been giving very accurate
results (Nagasaka et al., 2002; Benson et al., 2001; Noguchi et al., 2002). Gyros have
been widely used for inclination measurement (Mizushima et al., 2002). Fiber optic gyro
(FOG) has been observed to give the best performance (Nagasaka et al., 2002). At pres-
ent, gyros and inclinometers are available together as inertial measurement units (IMU)
for pitch, roll, and yaw, and linear velocity measurements. With the combination of RTK
GPS and FOG, an accuracy of ±5 cm (2.0 in) has been achieved (Noguchi et al., 2002).
GPS cannot be used alone for positioning in citrus applications as it gives errors when the
vehicle moves under tree canopies.
In addition to sensing global positions, the vehicle must be able to detect local obsta-
cles that may impede the path. Several sensing technologies have been explored for this
task. Ultrasonic sensors can map tree canopies while traveling at speeds of 1.8 m s
-1
(5.91 ft/s); measurement accuracy is better at lower speeds (Iida and Burks, 2002). The
development of machine vision guidance techniques has become a very attractive sens-
ing alternative, especially when combined with other proximity based sensors (Benson
et al., 2001; Zhang et al., 1999). They have proven to be reliable in several row-crop
