IEEE Robotics & Automation Magazine - September 2021 - 53

well as on the modes that can be profitably used to grasp
them. Picking tasks, the vast majority of which are currently
executed by human operators, can be classified into two
main categories: those that can be accomplished with one
hand and those that must be accomplished with two hands.
The problem of picking objects with a single end effector
has been extensively treated in the literature and in previous
articles of ours [42], [15]. The present article focuses
on the problem of picking objects that humans cannot pick
with one hand. Notably, the solution proposed here will
accomplish both categories of picking tasks. In the following,
a list of items that represents 40% of the volume of a
food warehouse is reported together with their main features.
Note that the food and beverage segment is one of
the market segments most affected by the e-commerce revolution,
which today allows customers to have their shopping
bags delivered directly to their homes. In the next
sections, we consider the functional requirements a picking
system should satisfy in order to work in such a warehouse
profitably. Finally, we describe the main challenges to be
tackled in the design and realization of this device.
Objects
The complete list of objects to be picked, along with their
size and weight, is reported in Figure 2. The objects can be
grouped into two sets depending on their shape: boxes or
cylinders. For the boxes, the values for the length (L),
height (H), and width (W) expressed in centimeters are
reported, while for the cylinders, the values for the diameter
(D) and height (H) are listed.
Functional Requirements
A list of functional requirements that a picking system
should match is reported in Table 1. It is important to stress
that these should not be taken as absolute values for every
intralogistics company but, rather, for operators that manage
a set of objects comparable to the one reported in Figure
2. These requirements are grouped into key
performance areas and indicators. Their quantitative value
should be considered as a target for the picking system.
Productivity performance indicators have been computed
based on the fact that such a robot would be economically
sustainable if it were able to perform three picking
movements every minute (corresponding to 180 picks per
hour). Given that a Euro-pallet (80- × 120-cm base piled up
to 1.5 m high) may contain up to 627 of the smallest or 115
of the largest among the objects reported in Figure 2, the
time needed to empty a pallet can be evaluated in 209 or
38 min, respectively. The average of these two values gives
the productivity performance indicator reported in Table 1.
The picking success rate takes into account the grasping
system without considering the vision system.
Challenges
The main challenges of the picking phase can be identified
as follows:
●
Boxes are often very close to each other, and the two
opposite sides, which are the most desirable for a reliable
and robust grasp, are usually not easily accessible. Hence,
to be properly handled, such boxes should first be moved
to guarantee that two opposite faces are accessible and
then picked.
●
Some items do not have a top surface, or the top surface
may not be suitable for grasping the object. These objects
cannot be grasped with vacuum grippers.
●
The bottom side of some objects is recessed under the
upper side of the objects that they lie under-or, more
generally, they cannot slide. This means that the objects
can translate only along the vertical direction or rotate
about a horizontal axis.
Human Picking Skills
There is no systematic method to synthesize all the requirements
listed in the section " Problem Definition " ; one of the
reasons is that it would
simultaneously involve the
codesign of the robot's
structure, planning, and
control. Hence, we ob -
served skilled human
operators at a food warehouse
during the exe -
cution of manipulation
tasks when picking the
objects listed in Figure 2.
That is, we recorded two
human operators from a
food warehouse while
performing the picking
action. Each picking action was repeated three times. These
live observations and the analysis of the video recordings
led to two lessons learned.
1) Bi-manual manipulation has a crucial role in picking operations
since humans use both hands to manipulate and handle
objects. In the majority of tasks, one hand is used to
move the object, and the other hand is used as a support.
Picking tasks, the vast
majority of which are
currently executed by
human operators, can be
classified into two
main categories.
2) The strategies human operators use to pick items are
classifiable into three main categories, as depending on
object shape and form, as depicted in Figure 3: rotation
about the horizontal axis, rotation about the vertical axis,
and sliding.
Rotation About the Horizontal Axis
In the case of thin boxes, i.e., H > W, H > L, and cylindrical
objects or if the support surface of an object cannot slide,
the operators use one hand to rotate the goods about a horizontal
axis and to put the object on the supporting hand
[see Figure 3(a)-(d)].
Rotation About the Vertical Axis
For thick boxes (H < W, H < L) with no constraints at the
base, the horizontal rotation is not convenient because of
SEPTEMBER 2021 * IEEE ROBOTICS & AUTOMATION MAGAZINE *
53
IEEE Robotics & Automation Magazine - September 2021

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - September 2021
