Next Article in Journal
Incorporation of Structural Similarity Index and Regularization Term into Neighbor2Neighbor Unsupervised Learning Model for Efficient Ultrasound Image Data Denoising
Previous Article in Journal
Assessing Ambisonics Sound Source Localization by Means of Virtual Reality and Gamification Tools
Previous Article in Special Issue
Optimization Method of the Clamping Force for Large Cabin Parts
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Applied Sciences
Volume 14
Issue 17
10.3390/app14177987
applsci-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

A Parcel Transportation and Delivery Mechanism for an Indoor Omnidirectional Robot

by
Elena Rubies
,
Ricard Bitriá
and
Jordi Palacín
*
Automation and Robotics Laboratory (ARL), Universitat de Lleida, 25001 Lleida, Spain
*
Author to whom correspondence should be addressed.
Appl. Sci. 2024, 14(17), 7987; https://doi.org/10.3390/app14177987
Submission received: 7 June 2024 / Revised: 8 August 2024 / Accepted: 3 September 2024 / Published: 6 September 2024
(This article belongs to the Special Issue New Insights into Intelligent Robotics)
Browse Figures
Versions Notes

Abstract

:
Parcel transportation is a task that is expected to be highly automated with the development of application-specific mobile robots. This paper presents the design and implementation of a parcel transportation and delivery mechanism aimed at converting a general-purpose indoor omnidirectional robot into an indoor delivery robot. The design of this new mechanism has considered the best placement in the robot and the limitation of not exceeding the original robot diameter. The mechanism consists of a basket with a lid that allows for the manual loading and automatic unloading of parcels. Despite the space limitations imposed by the general-purpose robot design, the designed mechanism can transport up to 90% of the packages received in an educational building. The mechanism was empirically validated by conducting 125 static manual loading experiments, 150 static unloading experiments, and 50 complete parcel delivery experiments. Results show that the delivery robot can efficiently deliver 78% of the total packages received in the building: envelopes, very small parcels, and small parcels. In the case of medium parcels, the delivery was unsuccessful in 30% of cases, in which the parcel did not properly slide out of the basket.

1. Introduction

Growths in e-commerce have consistently risen worldwide in the last decade, with a global increase of 6.5% in 2022 relative to 2021 [1]. Worldwide, 19% of all retail sales came from e-commerce during the same period [1]. The increment in online sales has resulted in the intensification of parcel deliveries [2,3] which is a challenge for parcel service providers and city logistics [4]. Currently, delivery vehicles are responsible for up to 20% of vehicular congestion and almost 60% of total emissions [5], operate inefficiently, and accumulate unnecessary mileage [6]. Indeed, the main challenges faced in last-mile delivery are sustainability, costs, time pressure, an increasing volume, and an aging workforce [7]. Since modern city logistics should be efficient and cause as few negative externalities as possible [8], conventional delivery based on vans is not considered a viable option for the future [9]. Hence, efforts on parcel shipments in urban areas are focused on reducing traffic volumes and proposing alternative delivery concepts. For instance, researchers are evaluating several alternative delivery options to identify efficient delivery systems for urban areas [10,11]. Among the innovative ideas, the benefits of using robots for last-mile delivery is a topic of interest [12].
Autonomous mobile robots have high capabilities such as autonomous trajectory planning, dynamic response to changes in the environment, and direct machine-to-machine interaction [13]. Therefore, autonomous robots can automate repetitive actions such as the transportation and delivery of parcels [14] because they are able to compute and implement efficient sequences of actions to transport and deliver packets from a pick-up to a drop-off point [15]. So far, the main research trend related to delivery robots has been focused on outdoor last-mile delivery [14,15,16,17,18]. Specifically, last-mile delivery refers to the logistical operations involved in delivering parcels to customers’ households [19]. While last-mile delivery is considered an outdoor problem that deals with transportation from depots to customers’ homes (single-family houses) [20], the problem rests unsolved for residential multi-story buildings with many apartments or offices per floor. Despite the requirements of robotic delivery systems operating outdoors and indoors are different, some works extrapolate the proposals and conclusions from outdoor parcel delivery works to applications related with indoor parcel delivery [21,22].
The scientific literature on indoor delivery is usually focused on the design of the robot or the algorithms used to navigate within a building [23]. For example, Abrar et al. [24] presented an autonomous delivery robot to prevent the spread of infectious diseases in product delivery during the coronavirus pandemic. The package was transported inside a password-protected container, and when the robot reached its destination, it waited for the customer to unlock the container and extract the product. Kim et al. [25] developed an indoor delivery robot for multi-floor environments. The system included a parcel box with a small door that remained closed during transportation, and when the robot arrived at the destination, if the recipient confirmed that they would receive the parcel, the robot opened the door, and the parcel could be picked up. In case the consignee did not respond, the robot moved to the next destination. In this case, parcel loading into the robot was automatic: the robot went to a hub with a manipulator that put the parcels into the mobile robot. Limosani et al. [26] proposed three heterogeneous mobile platforms that cooperated among themselves in domestic, condominium, and outdoor environments in order to transport items from outdoors to consignees’ apartments and vice versa. Transportation from outdoors to indoors was continuous thanks to a mechanism with rollers in the outdoor and condominium platforms that conveyed parcels from the enclosed box of the outdoor robot to the open tray in the condominium robot. Park et al. [27] developed a delivery robotic platform with a stair-climbing ability. In this case, the delivery module was on the top of the robot and could be changed depending on the transported cargo: an enclosed cool box to transport food, an open basket for large items, or simply flexible bands to securely fix cargo of various sizes. The system could also be used as an electric wheelchair. Muramatsu et al. [28] converted a stroller into a mobile delivery robot for indoor environments. The chair of the stroller was used as an open basket for parcel transportation, and the loading and unloading of parcels was manually performed by the sender and the recipient. Diddeniya et al. [29] presented an office assistant robot to deliver documents or parcels between office members. In this case, the mail was transported in a tray, and users had to manually drop it off and pick it up. Siao et al. [30] implemented a fully automatic meal delivery system with a robotic arm to put cups on tables without human intervention, and Lu et al. [31] included a cargo compartment for storing parcels in a robot, but they did not elaborate on this. Similarly, Naeem et al. [32] described the design of a non-holonomic autonomous delivery robot without detailing the system implemented to load and unload parcels. Jean et al. [33] deployed an office delivery robot with two manipulator arms with a gripper to facilitate document delivery tasks and also human–robot interactions, but they did not specify the transportation and delivery process. Choi et al. [34] designed an indoor service platform to interact with employees with special needs in a café to provide delivery services using a LIDAR sensor and two actuators to go to the target location, deliver a drink, and return. In a similar direction, devices such as the CIoT-Robot from Cao et al. [35] have tailored dispensing mechanisms aimed at delivering specific products such as medicines.
Other aspects of the delivery, such as the stabilization of the transportation compartment, have been addressed by Vikhe et al. [36] using a gyroscope and a motor on a four-legged delivery robot, and by Kim et al. [37] using a robotic arm for this stabilization.
As a summary, in the scientific literature, the design of the transportation and delivery mechanism is not addressed in detail, and parcel loading and unloading procedures have been automatized in very few proposals [25,26,30].
The new contribution of this paper is the design and implementation of a transportation and delivery mechanism with manual loading and automatic unloading capabilities. The aim of this mechanism is the conversion of a general-purpose omnidirectional mobile robot into an indoor delivery robot ready to operate inside a multi-story building. The design of the transportation and delivery mechanism allows the following operation: (1) the packages are manually received in the reception of the building, (2) a delivery robot is called, (3) one package is manually loaded onto the transportation mechanism of the robot, (4) the destination of the package (inside the building) is manually selected, (5) the robot moves autonomously from the reception point to the destination point, (6) the delivery robot automatically unloads the package into a predefined delivery area without requiring human attention, and (7) the delivery robot returns to its charging point. In case the route or the delivery cannot be completed, the robot returns the package to the reception.
This paper is structured as follows: Section 2 describes the materials and methods used. Section 3 presents the design of the transportation and delivery mechanism. Section 4 describes the implementation details of the transportation and delivery mechanism. Section 5 presents the experimental results obtained with the mechanism. Section 6 discusses the results obtained. Finally, Section 7 provides the conclusions of this paper and describes possible future works.

2. Materials and Methods

The main material used in this work is an existing omnidirectional mobile robot developed by the Automation and Robotics Laboratory at the Universitat de Lleida [38]. Figure 1 shows the robot, which is based on a tall (1.75 m) and thin central structure that supports a screen used as a robot face and two arms with 4 degrees of freedom each. The motion system of the mobile robot provides omnidirectional mobility capabilities [39,40], so the robot can rotate and translate simultaneously, being able to maneuver in small spaces. The motion system of the robot has a diameter of 0.56 m, which defines a cylindrical collision-free space around the robot. The omnidirectional mobile robot requires a navigable area with a diameter of 0.66 m of free space around the robot in order to be able to move and follow any trajectory combining translation and rotation. The unrestricted use of arm gestures may require a cylindrical collision-free space around the robot with a diameter of up to 1.50 m.
The mobile robot performs autonomous navigation. It creates a map of the environment using Simultaneous Location and Mapping (SLAM) based on the Iterative Closest Point (ICP) algorithm [41,42] and follows any path defined in the map by using the A* algorithm [43]. The data required by these processes are gathered by a 2D LIDAR (Hokuyo UTM-30LX, Osaka, Japan) and an RGB-D camera (Creative Senz3D, Tokyo, Japan). The LIDAR is located in the robot base, while the RGB-D camera is ground-oriented to detect obstacles in front of the robot, holes in the floor or stairs. Two supplementary Creative Senz3D RGB-D cameras provide additional information of the environment around the robot: one camera is in the upper part of the robot, above the screen, while the other is 1 m from the floor. Both cameras are aimed at detecting and discriminating between adults and children in front of the robot. The robot also includes a set of passive infrared (PIR) sensors to detect human activity. The location of all these sensors is highlighted in Figure 1.
The mobile robot was originally designed as an assistance tool [38], but it has also been used in other applications, for instance as a walk-helper device [44]. Thanks to its versatile motion system, parcel delivery is regarded as a potential application. The mobile robot also includes a passive suspension system [45] that reduces the transmission of vibrations generated by the wheels, making it suitable for transporting delicate packages.

3. Design of the Transportation and Delivery Mechanism

The design of the transportation and delivery mechanism for the existing mobile robot involves the following tasks: definition of its requirements [46], study of the type of parcels that it should carry, assessment of its location in the robot, and design.

3.1. Requirements of the Transportation and Delivery Mechanism

The main requirement of the transportation and delivery mechanism is to allow the selected mobile robot to transport and deliver parcels inside an office building. The mechanism must consist of a basket with a lid and allow the manual loading of the parcels in the robot and their automatic unloading. The assumption is that the transportation company delivers the parcels at the reception of the building, so the robot does not go outside to the street. Considering this, the transportation and delivery process should be as follows: The janitor loads one parcel at a time onto the transportation and delivery mechanism of the robot and defines the consignee or destination. Then, the robot autonomously plans the path and travels to the delivery destination. Finally, the robot automatically unloads the parcel in a specific delivery zone and returns to the starting point to pick up the next parcel. In this paper, it is assumed that the building has automatic doors that can be remotely opened by the robot [47], so the problem of manually opening and closing doors has not been addressed.

3.2. Requirements for the Automatic Unloading of Parcels

The automatic procedure proposed in this work to automatically unload parcels requires the definition of a delivery destination for the robot combined with the creation of a delivery zone for the parcels. For simplicity, the delivery zone is a tray defined over a table or desktop. The dimensions of the tray must correspond to the maximum size of the parcels delivered. The delivery destination defines an exact mobile robot location and orientation so that the transportation and delivery mechanism is aligned to unload the parcel onto the delivery zone no matter if the consignee is there or not.

3.3. Requirements for Robot Compatibility

As stated previously, the original omnidirectional mobile robot used in this work to create a delivery robot has a diameter of 0.56 m. The base of the robot contains the motion system and the control system. The upper part of the original robot has free space areas that can be used to place a transportation and delivery mechanism. The main requirements for robot compatibility are as follows: (1) the transportation zone in the robot cannot block the field of view of the original onboard sensors (LIDAR, RGB-D cameras, PIR sensors) and (2) the transportation zone cannot exceed the current diameter of the robot.
Therefore, any new structure attached to the robot must be inside the cylinder defined by its current diameter to take advantage of its already developed control system. Figure 2a represents the diameter of the robot, the free navigable area required by the robot to move (blue circle and line), and the free space remaining when going through a conventional 0.8 m wide door [48]. Figure 2a shows the mobile robot going through the door facing forwards, although, as an omnidirectional robot, it can go through the door in any orientation. Figure 2b highlights the specific motion performances provided by the omnidirectional robot, which can change its orientation without maneuvering while approaching a planned destination. These motion performances allow the robot to maintain a constant velocity during all the trajectories, minimizing the time required to arrive at the destination defined by a position and a robot orientation.

3.4. Analysis of the Parcels Received in an Educational Building

The parcels received in the Polytechnic School of the Universitat de Lleida (Spain) have been analyzed to guide the design of the transportation and delivery mechanism. This small-sized five-story educational building has 174 possible delivery destinations. The parcel analysis was conducted for 4 months by the staff of the building, from July to November 2023, inclusively, with the exception of the holiday period in August. Prior to starting the analysis, the parcels received were pre-screened to define some representative sizes and categories (Table 1): small envelope (SE), large envelope (LE), very small parcel (VSP), small parcel (SP), medium parcel (MP), large parcel (LP), and very large parcel (VLP). During the four months of the analysis, a total of 1070 envelopes and 727 parcels were received and classified into the nearest size category. This was documented by the janitors of the building that filled out a daily questionnaire with the size and classification of the parcels received. Table 1 shows the categories, and the total share of the parcels received [49,50]. Large envelopes (LEs) were the most frequently received (36%), followed by small envelopes (SEs) and small and medium parcels (SPs and MPs). In general, the share of the parcels received in a specific building depends on the main activities carried out [49,50], the period of time, the existence of holiday periods, and the existence of special purchase days. The aim of this analysis was to have a picture of the parcels received in a building where a variety of technological activities take place: mathematical analysis, electronic design and implementation, robotics, and mechanical design and implementation. The dimensions of the transportation and delivery mechanism will take the results of this analysis into account to carry as many parcels as possible.

3.5. Location of the Transportation Area in the Robot

This section presents the alternatives considered to include the parcel delivery functionality in the existing mobile robot (see Figure 1). Table 2 presents the possible locations of the transportation zone in the robot [51,52]. Each row presents one alternative assessed. The first column labels each one with a letter; the second column includes a graphical representation of the transportation zone represented; the third column represents the maximum size reachable in this configuration (limited by the diameter of the robot); and the fourth and fifth columns highlight the major pros and cons of each proposal.
The volume of the robot was used to define the dimensions of the transportation zones. The maximum sizes defined in Table 2 are inside the diameter of the robot base and limited by its central mast. The height for options A and B is half the distance between the LIDAR and the robot shoulders. The height for options C and D is half the distance between the top of the robot base and the robot shoulders.
The transportation zone alternatives were analyzed considering the transportation and delivery requirements. It was considered that the overriding characteristics of the transportation zone were the lack of interferences with the sensors and its height. Pondering these aforementioned aspects, options A and B were discarded because they interfere with the field of view of the 3D cameras and the sensors (see the location of the sensors in Figure 1). Option D was discarded because the transportation zone is at a height lower than a table. Hence, option C, locating the transportation zone behind the robot and in the upper half of its body, was regarded as the best alternative to include a transportation and delivery mechanism. According to the size of the transportation area defined in option C, the delivery zone placed over a table was defined as a tray of 0.4 m x 0.4 m with a side wall of 0.04 m.
As a summary, the positive features of the selected location are that the transportation zone does not block any onboard sensor of the robot, and its height is adequate to unload parcels onto a table. Alternatively, the main drawback of this proposal is the asymmetric distribution of the weight in the robot, which may require adjustments when using a passive suspension [45].

3.6. Design of the Transportation and Delivery Device

The design of the transportation and delivery device of the robot was based on the information on the types of envelopes and parcels received. A parallelepiped container resembling a basket with a lid was designed to fit the parcels. As stated before, the dimensions of the container were defined to fit in the most frequently received parcels [53] while not jutting out from the radius defined by the robot base. The selected dimensions were 0.34 W m × 0.12 L m × 0.22 H m, which allowed the accommodation of 90% of the mail received (categories SE, LE, VSP, SP, and MP, Table 1). The design was developed based on the assumption that large envelopes (LEs) and medium-sized parcels (MPs) can be transported with the lid open, but the large (LPs) and very large parcels (VLPs) will not fit in the transportation basket and must be delivered manually by the staff of the building.
Inspired by the Occam’s razor [54], the new mechanism was composed of four basic parts: a basket, a lid, and two static parts fixed to the main body of the robot. Both the basket and the lid must be articulated in order to automatically load and unload parcels. One of the static parts allocates a servomotor that holds and rotates the lid (Figure 3), and the other static part allocates a servomotor that holds and rotates the basket (Figure 4). In this design, the servomotors used to rotate the lid and the basket are low-cost 8120 MG digital servos, which provide a rotation angle of up to 270° and a torque of 2.2 Nm at 8.4 V. This low-cost servomotor uses an internal digital controller that provides a very high torque while keeping a minimum standard size (0.04 m × 0.02 m × 0.04 m). Both servomotors are controlled by an electronic board providing two digital PWM signals. The cables of the mechanism are hidden inside the main body of the robot, while the electronic board controlling the servomotors is fitted under the piece that holds the basket. The reference positions of the servomotors controlling the lid and the basket must be calibrated during the assembly of the device. Figure 3a shows the transportation position of the lid. In this position, the servomotor does not block the lid, allowing its manual opening by the janitor to load a parcel inside the basket. Figure 3b shows the lid lifted 90° to allow basket rotation. Figure 4a shows the basket in the transportation position, and Figure 4b shows the basket starting a rotation. The basket does not require the application of any torque during the transportation of a parcel. The automatic unloading of a parcel requires the basket to rotate 130° to drop the parcel off at the delivery zone by the action of gravity. The adequacy of the servomotors used in the design will be validated experimentally.

3.7. Parcel Loading Procedure

The sequence for the janitor to manually load a parcel onto the basket is as follows: (1) manually open the lid and keep it open; (2) put the parcel inside the basket; and (3) close the lid.

3.8. Parcel Unloading Procedure

Automatic parcel unloading is achieved by the servos of the mechanism. The sequential actions are as follows: (1) lift the lid up and keep it open; (2) rotate the basket down to the delivery position, so that gravity makes the parcel slide out of the basket; (3) rotate the basket up to the transportation position; and (4) rotate the lid down to the transportation position.

4. Implementation

Figure 5 shows the implementation of the transportation and delivery mechanism in the robot. The basket, the lid, and the two static parts of the mechanism are 3D-printed with polylactic acid (PLA). Figure 5a shows the mechanism in the transportation position. Figure 5b shows the lid open and the basket rotated to deliver a parcel over the delivery zone (green tray). Finally, Figure 5c shows the placement of the electronic control board and the servomotor controlling the rotation of the basket, which are hidden under the fixed part of the mechanism that holds the basket.

5. Results

5.1. Experiments on Manual Parcel Loading

Figure 6 shows the sequence to manually load the basket of the mechanism. The steps represented, from left to right, are: manually open and hold the lid, introduce the parcel inside the basket, and manually close the lid. Additionally, the delivery robot also provides a procedure for semi-automatic parcel loading, activated by a button displayed on its tactile screen. When this button is pressed, the robot opens the lid (no manual action is needed), keeps this position until a parcel destination is selected on the screen of the robot, and then automatically closes the lid. Table 3 shows the results of 250 parcel loading experiments conducted with the delivery robot: 125 opening the lid manually and 125 requesting its automatic opening. All parcel loading experiments were successful.

5.2. Preliminary Experiments on Parcel Unloading

The first experiments on parcel unloading performed with the delivery robot revealed some functional problems, which are shown in Figure 7. For each type of mail, the image in the first column illustrates the moment in which the robot reached its delivery destination and was ready to unload the parcel. The second column shows the lid open and the basket in the delivery position. The third and fourth columns show the parcel still inside the basket after the unloading procedure.
In the case of delivering a small envelope (SE, Figure 7a), when the basket rotated to drop the envelope off, it slid along the wall of the basket, but, due to its low weight, it did not fall out. In the case of the large envelope transported horizontally (LE-H, Figure 7b), its width was larger than the width of the basket, and it remained blocked inside the basket because, due to its low weight, the force of gravity was lower than the friction with the walls of the basket. Alternatively, in the case of delivering a medium parcel transported horizontally (MP-H, Figure 7c), it slid along the wall of the basket until it contacted the delivery zone, but the parcel did not have enough space to slide out of it.
Applsci 14 07987 g007
Figure 7. Sequence of the failed experiments on parcel unloading: (a) with SE; (b) with LE, placed horizontally (LE-H); and (c) with MP, placed horizontally (MP-H).
Figure 7. Sequence of the failed experiments on parcel unloading: (a) with SE; (b) with LE, placed horizontally (LE-H); and (c) with MP, placed horizontally (MP-H).
Applsci 14 07987 g007
A person would solve this problem by shaking the basket, so the adopted solution was making the basket oscillate ±5° around the delivery position to facilitate parcel unloading from the basket. This improved parcel unloading procedure was defined as follows: (1) lift the lid up and hold it open; (2) start to rotate the basket down to the delivery position (at approximately 130° from the vertical transport position); (3) one second after the rotation of the basket has started, make an oscillation around the delivery position to facilitate the parcel to slide out of the basket: +5° for 0.1 s, −5° for 0.1 s, +5° for 0.1 s, and −5° for 0.1 s; (4) rotate the basket up to the transportation position; and (5) rotate the lid down to the transportation position.

5.3. Experiments on Automatic Parcel Unloading

This section shows the experiments on automatic parcel unloading conducted with the delivery robot. Table 4 shows the results of 25 automatic parcel unloading experiments conducted for each type of parcel category, while Figure 8 shows the evolution of one experiment of each parcel category. The image in the first column of Figure 8 illustrates the moment in which the delivery robot is ready to unload the parcel. The second column shows the lid open and the basket rotating down to drop the parcel off. The third column shows the parcel in the delivery area and the basket returning to the transportation position. The fourth column shows both the basket and the lid in the transportation position, so the robot is ready to return to the reception of the building.
In each experiment, the robot was statically placed in a delivery destination, with its back oriented to the delivery zone (see Figure 8). When an experiment was started, the robot performed the improved unloading procedure described in the previous section. The unloading experiments with the small envelope (SE) (Figure 8a), the large envelope (LE) transported horizontally (Figure 8b) and vertically (Figure 8c), the very small parcel (VSP) and the small parcel (SP) transported horizontally (Figure 8d) and vertically (Figure 8e) were successful thanks to the oscillation of the basket during the delivery.
In the case of unloading a medium parcel (MP), the success rate was 68% when the parcel is transported horizontally (Figure 8f) and 72% when the parcel was transported vertically (Figure 8g). Failures occur when the parcel gets stuck and does not slide out of the basket. Nevertheless, thanks to the application of the shaking procedure, the success rate in the unloading of medium parcels (MPs) was improved from 0% to an average of 70%.
Applsci 14 07987 g008
Figure 8. Sequence of experiments on automatic parcel unloading: (a) with SE; (b) with LE transported horizontally (LE-H) and (c) vertically (LE-V); (d) with SP transported horizontally (SP-H) and (e) vertically (SP-V); (f) with MP transported horizontally (MP-H) and (g) vertically (MP-V).
Figure 8. Sequence of experiments on automatic parcel unloading: (a) with SE; (b) with LE transported horizontally (LE-H) and (c) vertically (LE-V); (d) with SP transported horizontally (SP-H) and (e) vertically (SP-V); (f) with MP transported horizontally (MP-H) and (g) vertically (MP-V).
Applsci 14 07987 g008

5.4. Experiments on Parcel Transportation and Delivery

The operability of the mechanism was assessed in a first set of complete parcel transportation and delivery experiments developed in a multi-story building. This first set of delivery experiments was performed only with small parcels (SPs). This section is focused on the evaluation of the experiments from the point of view of the mechanism for parcel transportation and delivery. The procedures used to control the mobile robot and implement all the steps required to complete a parcel transportation and delivery task will be explained in a separate work that will be published elsewhere.
Each parcel delivery experiment consisted of the following steps: (1) a parcel is received in the reception of the building; (2) the delivery robot is called so that it goes from its charging station to a predefined specific position at the reception; (3) the parcel is manually loaded onto the robot; (4) a delivery destination inside the building is selected by using the tactile screen of the robot; (5) the robot performs path planning [55] and starts the delivery; (6) if the navigation is successful, the robot arrives at the delivery destination, with its back oriented to the delivery zone; (7) then, the robot unloads the parcel; and (8) the robot waits some time in case the cardboard of the package needs to be returned back for recycling. After completing the delivery, (9) the robot performs path planning to return to the charging station, and (10) if the navigation is successful, the robot arrives at the charging station or at the reception of the building in the case of receiving another call to deliver a new package. In the cases in which the navigation to reach the destination is unsuccessful, the mobile robot is configured to return the parcel to the reception. Specifically, Table 5 reports the experimental results obtained in steps (3) in which a parcel is manually loaded onto the delivery robot, and (7) in which the delivery robot unloads the parcel at the delivery zone.
Table 5 shows the results of step (3) and step (7) evaluated in 25 complete transportation and delivery tasks. Figure 9 shows some images of the robot completing step (7): entering an office, arriving at the delivery destination with its back oriented to the delivery zone, automatically unloading the parcel, waiting some time to collect any waste from the parcel, and returning to the charging station or the reception of the building.
In all the delivery tasks, the manual loading of the parcel was always successful, while the automatic unloading of the parcel was successful in 22 experiments in which the robot arrived at its delivery destination. However, in three of the delivery experiments, the robot was not able to arrive at the delivery destination because of a navigation problem, so the parcel was kept loaded in the robot and returned to the reception of the building. These unsuccessful deliveries were not caused by a failure of the parcel transportation and delivery mechanism, so they have not been considered as a failure from the point of view of automatic unloading at the delivery destination (see Table 5).
Applsci 14 07987 g009
Figure 9. Sequence of one experiment on parcel transportation and delivery. The arrows represent robot movements (yellow) and parcel delivery (blue).
Figure 9. Sequence of one experiment on parcel transportation and delivery. The arrows represent robot movements (yellow) and parcel delivery (blue).
Applsci 14 07987 g009

6. Discussion

This paper presents the design and implementation of a parcel transportation and delivery mechanism aimed at converting a general-purpose indoor omnidirectional robot into an indoor delivery robot.
In general, the scientific literature on indoor delivery robots is usually focused on the control of the robot [23] without paying much attention to the transportation mechanism or the delivery of the packages transported. The design of this new transportation and delivery mechanism considered the best placement in the robot and the limitation of not exceeding the diameter of the original robot to which it was attached. The final design of the mechanism was placed behind the robot and in the upper half of its body. The design consists of a basket with a lid, articulated with servomotors, which allow the manual loading and the automatic unloading of parcels.
Compared with the transportation mechanism proposed by Kim et al. [25], this new mechanism design is able to transport and automatically unload parcels onto predefined delivery areas without human intervention. Compared with the delivery mechanism proposed by Siao et al. [30], this new mechanism is not limited to the delivery of cups, and no additional stabilization is needed thanks to the passive suspension embedded in the original omnidirectional mobile robot. Compared with the robotic delivery system proposed for home apartments by Limosani et al. [26], this new mechanism is designed to operate in multi-story office buildings, while both proposals are conceived to allow for the collection and return of any envelope or package garbage.
The design of the transportation and delivery mechanism considered the size of 1797 packages received within four months in a five-story educational building with 174 possible delivery destinations. The analysis of the size of the packages received showed that the parcels most frequently received were large envelopes (36%), followed by small envelopes (23%) and small (14%) and medium (12%) parcels. Based on this information, the size of the basket of the mechanism was able to accommodate up to 90% of the packages received in the building.
The design of the mechanism allows the following delivery operation: (1) the packages are manually received in the reception of the building, (2) the delivery robot is called, (3) one package is manually loaded onto the transportation mechanism of the robot, (4) the destination of the package in the building is manually selected, (5) the robot moves autonomously from the reception point to the destination point, (6) the delivery robot automatically unloads the package into a predefined delivery area, and (7) the delivery robot returns to its charging point. In cases where the route or the delivery cannot be completed, the robot returns the package to the reception.
The transportation and delivery mechanism was empirically validated by conducting 125 static manual loading experiments, 150 static unloading experiments, and 50 complete parcel delivery experiments. All static manual parcel loading experiments were successfully completed, but preliminary parcel unloading experiments revealed unloading problems because some parcels did not slide out of the basket. This problem was solved by considering what a person would do, defining an oscillation (shake) around the delivery position of the basket.
After this procedure enhancement, results showed that the delivery robot can efficiently deliver small envelopes, large envelopes, very small parcels, and small parcels, which represent 78% of the total packages received in the building. In the case of medium-sized parcels, the delivery was unsuccessful in 30% of the cases, in which the parcel did not slide out of the basket properly. Specifically, the delivery experiments with medium-sized parcels were successful in 68% of the tests when the parcels were horizontally loaded and in 72% of the tests when the parcels were vertically loaded. This is because their size was similar to the size of the basket, and, in some attempts, they did not fully slide out of the basket despite the additional shake introduced. These results can be improved in future implementations with the incorporation of a mechanism to push the packages out of the basket. Finally, a first set of complete dynamic delivery experiments performed with small parcels validated the transportation and delivery mechanism in real operation conditions inside a multi-story building.

7. Conclusions

The main conclusion of this work is that an omnidirectional mobile robot equipped with a low-cost transportation and delivery mechanism is capable of efficiently performing parcel delivery inside buildings. The main challenge was unloading large packages from the transportation basket. Future work will be focused on user interaction and on the control of the mobile robot to complete parcel delivery in crowded multi-story buildings such as hospitals.

Author Contributions

Formal analysis, E.R.; investigation, E.R., R.B. and J.P.; methodology, J.P.; resources, R.B.; validation, E.R. and R.B.; writing—original draft, E.R.; writing—review and editing, J.P. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by Departament de Recerca i Universitats de la Generalitat de Catalunya: AGAUR FI SDUR 2022, and Ministerio de Ciencia, Innovación y Universidades: FPU 2022/00526.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The original contributions presented in the study are included in the article, further inquiries can be directed to the corresponding authors.

Conflicts of Interest

The authors declare no conflicts of interest. The funders had no role in the design of this study; in the collection, analyses, or interpretation of data; in the writing of this manuscript; or in the decision to publish the results.

References

  1. Coppola, D. Annual Retail E-Commerce Sales Growth Worldwide from 2017 to 2027. Available online: https://www.statista.com/statistics/288487/forecast-of-global-b2c-e-commerce-growth/ (accessed on 19 September 2023).
  2. Leyerer, M.; Sonneberg, M.-O.; Heumann, M.; Breitner, M.H. Decision Support for Sustainable and Resilience-Oriented Urban Parcel Delivery. EURO J. Decis. Process. 2019, 7, 267–300. [Google Scholar] [CrossRef]
  3. Chen, M.-C.; Wu, P.-J.; Hsu, Y.-H. An Effective Pricing Model for the Congestion Alleviation of E-Commerce Logistics. Comput. Ind. Eng. 2019, 129, 368–376. [Google Scholar] [CrossRef]
  4. Cárdenas, I.; Beckers, J.; Vanelslander, T.; Verhetsel, A. Spatial Characteristics of Failed and Successful E-Commerce Deliveries in Belgian Cities. In Proceedings of the 6th International Conference on Information Systems, Logistics and Supply Chain, Bordeaux, France, 1–4 June 2016. [Google Scholar]
  5. Dablanc, L.; Diziain, D.; Levifve, H. Urban Freight Consultations in the Paris Region. Eur. Transp. Res. Rev. 2011, 3, 47–57. [Google Scholar] [CrossRef]
  6. Jiang, L.; Mahmassani, H.S. City Logistics: Freight Distribution Management with Time-Dependent Travel Times and Disruptive Events. Transp. Res. Rec. 2014, 2410, 85–95. [Google Scholar] [CrossRef]
  7. Boysen, N.; Fedtke, S.; Schwerdfeger, S. Last-Mile Delivery Concepts: A Survey from an Operational Research Perspective. OR Spectr. 2021, 43, 1–58. [Google Scholar] [CrossRef]
  8. Anderluh, A.; Hemmelmayr, V.C.; Rüdiger, D. Analytic Hierarchy Process for City Hub Location Selection—The Viennese Case. Transp. Res. Procedia 2020, 46, 77–84. [Google Scholar] [CrossRef]
  9. Li, L.; He, X.; Keoleian, G.A.; Kim, H.C.; De Kleine, R.; Wallington, T.J.; Kemp, N.J. Life Cycle Greenhouse Gas Emissions for Last-Mile Parcel Delivery by Automated Vehicles and Robots. Environ. Sci. Technol. 2021, 55, 11360–11367. [Google Scholar] [CrossRef] [PubMed]
  10. Himstedt, B.; Meisel, F. Parcel Delivery Systems for City Logistics: A Cost-Based Comparison between Different Transportation Technologies. J. Logist. Res. 2023, 16, 3. [Google Scholar] [CrossRef]
  11. Mohammad, W.A.; Nazih Diab, Y.; Elomri, A.; Triki, C. Innovative Solutions in Last Mile Delivery: Concepts, Practices, Challenges, and Future Directions. Supply Chain Forum: Int. J. 2023, 24, 151–169. [Google Scholar] [CrossRef]
  12. Ostermeier, M.; Heimfarth, A.; Hübner, A. The Multi-Vehicle Truck-and-Robot Routing Problem for Last-Mile Delivery. Eur. J. Oper. Res. 2023, 310, 680–697. [Google Scholar] [CrossRef]
  13. Fragapane, G.; de Koster, R.; Sgarbossa, F.; Strandhagen, J.O. Planning and Control of Autonomous Mobile Robots for Intralogistics: Literature Review and Research Agenda. Eur. J. Oper. Res. 2021, 294, 405–426. [Google Scholar] [CrossRef]
  14. Kim, J.; Moon, H.; Jung, H. Drone-Based Parcel Delivery Using the Rooftops of City Buildings: Model and Solution. Appl. Sci. 2020, 10, 4362. [Google Scholar] [CrossRef]
  15. Galindo, C.; Fernández-Madrigal, J.-A.; González, J.; Saffiotti, A. Robot Task Planning Using Semantic Maps. Robot. Auton. Syst. 2008, 56, 955–966. [Google Scholar] [CrossRef]
  16. Poeting, M.; Schaudt, S.; Clausen, U. A Comprehensive Case Study in Last-Mile Delivery Concepts for Parcel Robots. In Proceedings of the 2019 Winter Simulation Conference (WSC), National Harbor, MD, USA, 8–11 December 2019; pp. 1779–1788. [Google Scholar]
  17. Samouh, F.; Gluza, V.; Djavadian, S.; Meshkani, S.; Farooq, B. Multimodal Autonomous Last-Mile Delivery System Design and Application. In Proceedings of the 2020 IEEE International Smart Cities Conference (ISC2), Piscataway, NJ, USA, 28 September–1 October 2020; pp. 1–7. [Google Scholar]
  18. Gao, F.; Cheng, Y.; Gao, M.; Ma, C.; Liu, H.; Ren, Q.; Zhao, Z. Design and Implementation of an Autonomous Driving Delivery Robot. In Proceedings of the 2022 41st Chinese Control Conference (CCC), Hefei, China, 25–27 July 2022; pp. 3832–3837. [Google Scholar]
  19. Langevin, A.; Mbaraga, P.; Campbell, J.F. Continuous Approximation Models in Freight Distribution: An Overview. Transp. Res. Part B Methodol. 1996, 30, 163–188. [Google Scholar] [CrossRef]
  20. Muñoz-Villamizar, A.; Velazquez-Martínez, J.C.; Caballero-Caballero, S. A Large-Scale Last-Mile Consolidation Model for e-Commerce Home Delivery. Expert Syst. Appl. 2024, 235, 121200. [Google Scholar] [CrossRef]
  21. Hutter, M.; Gehring, C.; Lauber, A.; Gunther, F.; Bellicoso, C.D.; Tsounis, V.; Fankhauser, P.; Diethelm, R.; Bachmann, S.; Bloesch, M.; et al. ANYmal—Toward Legged Robots for Harsh Environments. Adv. Robot. 2017, 31, 918–931. [Google Scholar] [CrossRef]
  22. Castillo, G.A.; Weng, B.; Zhang, W.; Hereid, A. Robust Feedback Motion Policy Design Using Reinforcement Learning on a 3D Digit Bipedal Robot. In Proceedings of the 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Prague, Czech Republic, 27 September–1 October 2021; pp. 5136–5143. [Google Scholar]
  23. Sun, Y.; Guan, L.; Chang, Z.; Li, C.; Gao, Y. Design of a Low-Cost Indoor Navigation System for Food Delivery Robot Based on Multi-Sensor Information Fusion. Sensors 2019, 19, 4980. [Google Scholar] [CrossRef] [PubMed]
  24. Abrar, M.M.; Islam, R.; Shanto, M.A.H. An Autonomous Delivery Robot to Prevent the Spread of Coronavirus in Product Delivery System. In Proceedings of the 2020 11th IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON), New York, NY, USA, 28–31 October 2020; pp. 0461–0466. [Google Scholar]
  25. Kim, T.; Kang, G.; Lee, D.; Shim, D.H. Development of an Indoor Delivery Mobile Robot for a Multi-Floor Environment. IEEE Access 2024, 12, 45202–45215. [Google Scholar] [CrossRef]
  26. Limosani, R.; Esposito, R.; Manzi, A.; Teti, G.; Cavallo, F.; Dario, P. Robotic Delivery Service in Combined Outdoor–Indoor Environments: Technical Analysis and User Evaluation. Robot. Auton. Syst. 2018, 103, 56–67. [Google Scholar] [CrossRef]
  27. Park, S.; Ahn, S.; Shin, J.; Kim, H.; Yang, J.; Kim, Y.; Lim, K.; Seo, T. WAVE: Last Mile Delivery Robotic Platform With Stair-Climbing Ability Via Four-Bar Linkage-Based Locomotion. IEEE/ASME Trans. Mechatron. 2024, 29, 313–323. [Google Scholar] [CrossRef]
  28. Muramatsu, S.; Takahashi, K.; Tomizawa, T.; Kudoh, S.; Suehiro, T. Development of the Robust Delivery Robot System with the Unknown Object in Indoor Environment. In Proceedings of the IECON 2013—39th Annual Conference of the IEEE Industrial Electronics Society, Vienna, Austria, 10–13 November 2013; pp. 8271–8276. [Google Scholar]
  29. AP Diddeniya, S.I.; Adikari, A.M.S.B.; Gunasinghe, H.N.; De Silva, P.R.S.; Ganegoda, N.C.; Wanniarachchi, W.K.I.L. Vision Based Office Assistant Robot System for Indoor Office Environment. In Proceedings of the 2018 3rd International Conference on Information Technology Research (ICITR), Moratuwa, Sri Lanka, 5–7 December 2018; pp. 1–6. [Google Scholar]
  30. Siao, C.-Y.; Lin, J.-W.; Chang, R.-G. The Design and Implementation of A Delivery System. In Proceedings of the 2020 IEEE Eurasia Conference on IOT, Communication and Engineering (ECICE), Yunlin, Taiwan, 23–25 October 2020; pp. 160–163. [Google Scholar]
  31. Lu, Y.; Liu, Y.; Bu, Q.; Lim, E.G.; Devaraj, R.; Zhang, Q.; Sun, J. An Autonomous Vehicle Platform for Parcel Delivery. In Proceedings of the 2023 28th International Conference on Automation and Computing (ICAC), Birmingham, UK, 30 August–1 September 2023; pp. 1–7. [Google Scholar]
  32. Naeem, M.; Ahmed, A.; Ayman, F.; Ahmed, M.; Adel, A.; Amr, Y.; Ashraf, M. Autonomous Delivery Rocker-Bogie Robot Using Gmapping Based SLAM Algorithm. Int. Res. J. Eng. Technol. (IRJET) 2022, 09, 1608–1619. [Google Scholar]
  33. Jean, J.-H.; Wei, C.-F.; Lin, Z.-W.; Lian, K.-Y. Development of an Office Delivery Robot with Multimodal Human-Robot Interactions. In Proceedings of the 2012 Proceedings of SICE Annual Conference (SICE), Akita, Japan, 20–23 August 2012; pp. 1564–1567.
  34. Choi, J.; Kim, M.; Kim, J.; Lee, W. Designing an Interactive Indoor Delivery Robot and Its Implications. In Robot Intelligence Technology and Applications 6; Kim, J., Englot, B., Park, H.-W., Choi, H.-L., Myung, H., Kim, J., Kim, J.-H., Eds.; Springer International Publishing: Cham, Switzerland, 2022; pp. 201–209. [Google Scholar]
  35. Cao, H.; Huang, X.; Zhuang, J.; Xu, J.; Shao, Z. CIoT-Robot Cloud and IoT Assisted Indoor Robot for Medicine Delivery; Atlantis Press: Dordrecht, The Netherlands, 2018; pp. 85–89. [Google Scholar]
  36. Vikhe, M.; Bothra, N.; Malaviya, N.; Kothari, M.; Koshti, R. Four Legged Parallel Manipulator for Autonomous Delivery Robot. In Proceedings of the 2023 International Conference on Device Intelligence, Computing and Communication Technologies, (DICCT), Dehradun, India, 17–18 March 2023; pp. 346–350. [Google Scholar]
  37. Kim, M.; Choi, D. Design and Development of a Variable Configuration Delivery Robot Platform. Int. J. Precis. Eng. Manuf. 2019, 20, 1757–1765. [Google Scholar] [CrossRef]
  38. Palacín, J.; Rubies, E.; Clotet, E. The Assistant Personal Robot Project: From the APR-01 to the APR-02 Mobile Robot Prototypes. Designs 2022, 6, 66. [Google Scholar] [CrossRef]
  39. Qian, J.; Zi, B.; Wang, D.; Ma, Y.; Zhang, D. The Design and Development of an Omni-Directional Mobile Robot Oriented to an Intelligent Manufacturing System. Sensors 2017, 17, 2073. [Google Scholar] [CrossRef] [PubMed]
  40. Crenganiș, M.; Breaz, R.-E.; Racz, S.-G.; Gîrjob, C.-E.; Biriș, C.-M.; Maroșan, A.; Bârsan, A. Fuzzy Logic-Based Driving Decision for an Omnidirectional Mobile Robot Using a Simulink Dynamic Model. Appl. Sci. 2024, 14, 3058. [Google Scholar] [CrossRef]
  41. Clotet, E.; Palacín, J. SLAMICP Library: Accelerating Obstacle Detection in Mobile Robot Navigation via Outlier Monitoring Following ICP Localization. Sensors 2023, 23, 6841. [Google Scholar] [CrossRef] [PubMed]
  42. Yokozuka, M.; Koide, K.; Oishi, S.; Banno, A. LiTAMIN: LiDAR-Based Tracking And Mapping by Stabilized ICP for Geometry Approximation with Normal Distributions. In Proceedings of the 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Las Vegas, NV, USA, 24 October 2020–24 January 2021; pp. 5143–5150. [Google Scholar]
  43. Li, Y.; Dai, S.; Shi, Y.; Zhao, L.; Ding, M. Navigation Simulation of a Mecanum Wheel Mobile Robot Based on an Improved A* Algorithm in Unity3D. Sensors 2019, 19, 2976. [Google Scholar] [CrossRef] [PubMed]
  44. Palacín, J.; Clotet, E.; Martínez, D.; Martínez, D.; Moreno, J. Extending the Application of an Assistant Personal Robot as a Walk-Helper Tool. Robotics 2019, 8, 27. [Google Scholar] [CrossRef]
  45. Moreno, J.; Clotet, E.; Tresanchez, M.; Martínez, D.; Casanovas, J.; Palacín, J. Measurement of Vibrations in Two Tower-Typed Assistant Personal Robot Implementations with and without a Passive Suspension System. Sensors 2017, 17, 1122. [Google Scholar] [CrossRef]
  46. Silvestri, P.; Zoppi, M.; Molfino, R. Dynamic Investigation on a New Robotized Vehicle for Urban Freight Transport. Simul. Model. Pract. Theory 2019, 96, 101938. [Google Scholar] [CrossRef]
  47. Palacín, J.; Bitriá, R.; Rubies, E.; Clotet, E. A Procedure for Taking a Remotely Controlled Elevator with an Autonomous Mobile Robot Based on 2D LIDAR. Sensors 2023, 23, 6089. [Google Scholar] [CrossRef] [PubMed]
  48. Ministerio de Transportes, Movilidad y Agenda Urbana Código Técnico de la Edificación (CTE). Documento Básico—Seguridad de Utilización y Accesibilidad (DB-SUA). Available online: https://www.codigotecnico.org/DocumentosCTE/SeguridadUtilizacionAccesibilidad.html (accessed on 15 May 2024).
  49. van Duin, J.H.R.; Wiegmans, B.W.; van Arem, B.; van Amstel, Y. From Home Delivery to Parcel Lockers: A Case Study in Amsterdam. Transp. Res. Procedia 2020, 46, 37–44. [Google Scholar] [CrossRef]
  50. Merkert, R.; Bliemer, M.C.J.; Fayyaz, M. Consumer Preferences for Innovative and Traditional Last-Mile Parcel Delivery. Int. J. Phys. Distrib. Logist. Manag. 2022, 52, 261–284. [Google Scholar] [CrossRef]
  51. Tian, Y.; Zhang, S.; Liu, J.; Chen, F.; Li, L.; Xia, B. Research on a New Omnidirectional Mobile Platform with Heavy Loading and Flexible Motion. Adv. Mech. Eng. 2017, 9, 1687814017726683. [Google Scholar] [CrossRef]
  52. Gao, Z.Q.; Chen, H.B.; Du, Y.P.; Wei, L. Design and Development of an Omni-Directional Mobile Robot for Logistics. Appl. Mech. Mater. 2014, 602–605, 1006–1010. [Google Scholar] [CrossRef]
  53. Ranjbari, A.; Diehl, C.; Dalla Chiara, G.; Goodchild, A. What Is the Right Size for a Residential Building Parcel Locker? Transp. Res. Rec. 2023, 2677, 1397–1407. [Google Scholar] [CrossRef]
  54. Standish, R.K. Why Occam’s Razor. Found Phys Lett 2004, 17, 255–266. [Google Scholar] [CrossRef]
  55. Palacín, J.; Rubies, E.; Bitriá, R.; Clotet, E. Path Planning of a Mobile Delivery Robot Operating in a Multi-Story Building Based on a Predefined Navigation Tree. Sensors 2023, 23, 8795. [Google Scholar] [CrossRef] [PubMed]
Applsci 14 07987 g001
Figure 1. Mobile robot with the location of its main sensors highlighted: LIDAR (blue), RGB-D cameras pointing at the ground (red) and to the front (green and yellow), and PIR sensors (orange).
Figure 1. Mobile robot with the location of its main sensors highlighted: LIDAR (blue), RGB-D cameras pointing at the ground (red) and to the front (green and yellow), and PIR sensors (orange).
Applsci 14 07987 g001
Applsci 14 07987 g002
Figure 2. Top view of the omnidirectional mobile robot and its free navigable area required to move (blue circle and lines): (a) passing through a door and (b) rotating while moving to approach a table. The green arrows depict the orientation of the frontal part of the robot.
Figure 2. Top view of the omnidirectional mobile robot and its free navigable area required to move (blue circle and lines): (a) passing through a door and (b) rotating while moving to approach a table. The green arrows depict the orientation of the frontal part of the robot.
Applsci 14 07987 g002
Applsci 14 07987 g003
Figure 3. Detail of the fixed part attached to the body of the robot, with the servomotor (red) that drives the lid: (a) transportation position (closed lid); (b) delivery position (open lid).
Figure 3. Detail of the fixed part attached to the body of the robot, with the servomotor (red) that drives the lid: (a) transportation position (closed lid); (b) delivery position (open lid).
Applsci 14 07987 g003
Applsci 14 07987 g004
Figure 4. Detail of the fixed part attached to the body of the robot, with the servomotor (red) that drives the basket: (a) basket in the transportation position; (b) basket rotating to deliver a parcel.
Figure 4. Detail of the fixed part attached to the body of the robot, with the servomotor (red) that drives the basket: (a) basket in the transportation position; (b) basket rotating to deliver a parcel.
Applsci 14 07987 g004
Applsci 14 07987 g005
Figure 5. Transportation and delivery mechanism implemented in the robot: (a) transportation position; (b) delivery position; (c) bottom view showing the placement of the electronic control board and the servomotor of the basket.
Figure 5. Transportation and delivery mechanism implemented in the robot: (a) transportation position; (b) delivery position; (c) bottom view showing the placement of the electronic control board and the servomotor of the basket.
Applsci 14 07987 g005
Applsci 14 07987 g006
Figure 6. Sequence of one experiment on manual parcel loading onto the delivery robot.
Figure 6. Sequence of one experiment on manual parcel loading onto the delivery robot.
Applsci 14 07987 g006
Table 1. Dimensions and share of the mail received.
Table 1. Dimensions and share of the mail received.
CategoryLabelMaximum Size: W (m) × L (m) × H (m)Total ParcelsShare (%)
Small envelopeSE0.11 × 0.23 × 0.00541723.21
Large envelopeLE0.26 × 0.36 × 0.0165336.34
Very small parcelVSP0.13 × 0.09 × 0.04824.56
Small parcelSP0.23 × 0.14 × 0.0825013.91
Medium parcelMP0.32 × 0.22 × 0.1320911.63
Large parcelLP0.39 × 0.28 × 0.191437.96
Very large parcelVLP0.50 × 0.30 × 0.30432.39
Table 2. Alternatives for the location of the transportation area.
Table 2. Alternatives for the location of the transportation area.
LocationGraphical
Representation		Maximum Size:
W (m) × D (m) × H (m)		Main ProsMain Cons
AApplsci 14 07987 i0010.53 × 0.22 × 0.42Transportation zone height.Interferences with frontal and ground cameras.
Weight distribution and stability.
BApplsci 14 07987 i0020.53 × 0.22 × 0.42Weight distribution and stability.Interferences with ground camera.
Transportation zone height.
CApplsci 14 07987 i0030.53 × 0.18 × 0.44Transportation zone height.
No camera interferences.		Weight distribution and stability.
DApplsci 14 07987 i0040.53 × 0.18 × 0.44No camera interferences.
Weight distribution and stability.		Transportation zone height.
Table 3. Results of the experiments on manual and semi-automatic parcel loading.
Table 3. Results of the experiments on manual and semi-automatic parcel loading.
Mobile RobotActionParcelExperimentsSuccess
StationaryManual loadSE25100%
LE25100%
VSP25100%
SP25100%
MP25100%
Requested semi-automatic loadSE25100%
LE25100%
VSP25100%
SP25100%
MP25100%
Table 4. Results of the experiments on automatic parcel unloading.
Table 4. Results of the experiments on automatic parcel unloading.
Mobile RobotActionParcelExperimentsSuccess
Stationary (placed in the delivery destination)Automatic unloadSE25100%
LE25100%
VSP25100%
SP25100%
MP-H, horizontal2568%
MP-V, vertical2572%
Table 5. Results of the experiments on mobile robot parcel transportation and delivery.
Table 5. Results of the experiments on mobile robot parcel transportation and delivery.
Mobile Robot(Step) ActionParcelExperimentsSuccess
Performing parcel transportation and delivery (3) Manual loadSP25100%
(7) Automatic unloadSP25 1100%
1 In 3 experiments, the robot could not reach destination and returned the parcel to the reception.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Rubies, E.; Bitriá, R.; Palacín, J. A Parcel Transportation and Delivery Mechanism for an Indoor Omnidirectional Robot. Appl. Sci. 2024, 14, 7987. https://doi.org/10.3390/app14177987

AMA Style

Rubies E, Bitriá R, Palacín J. A Parcel Transportation and Delivery Mechanism for an Indoor Omnidirectional Robot. Applied Sciences. 2024; 14(17):7987. https://doi.org/10.3390/app14177987

Chicago/Turabian Style

Rubies, Elena, Ricard Bitriá, and Jordi Palacín. 2024. "A Parcel Transportation and Delivery Mechanism for an Indoor Omnidirectional Robot" Applied Sciences 14, no. 17: 7987. https://doi.org/10.3390/app14177987

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop