border border

Unmanned Delivery System

Team members

Brendan Ong Yen Kie (EPD), Chia Wing Yee Rachel (EPD), Keane Hi Zhe Wei (EPD), Lum Kai Wen (EPD), Ling Zi Wei Rebecca (ISTD), Yee Yi Xian (ISTD)

Instructors:

Ye Ai, Cyrille Pierre Joseph Jegourel

Writing Instructors:

Rashmi Kumar

Teaching Assistant:

Congjian Lin

In Collaboration with:

 

 

 

The Problem: Delayed and Damaged Parcels

In today's e-commerce era, there has been an increased demand for online retail trade causing a higher demand for delivery services. However, many local companies are unable to cope with this increasing demand due to manpower shortages, resulting in late, lost, or damaged deliveries. This results in frustrated customers, and increased burden on already overworked delivery workers.

Current Delivery Process

The traditional delivery process is greatly reliant on manpower, where the collection, sorting and delivery are done by company employees. With a shortage of manpower, many choke points occur during the periods when the courier collects and delivers a parcel from the store and the sorting hub respectively.

Thus leading us to our aim:

By leveraging on a fleet of autonomous robotic delivery systems, how might we alleviate the issue of long delivery times due to the shortage of manpower, such that the parcel delivery process would be more efficient and require less human involvement.​

     

The Solution: Unmanned Delivery System



A Novel End-to-End Delivery System

The Broad Architecture

Our delivery system is made up of 4 main components:



These 4 systems work seamlessly with each other through a central server and RESTful APIs to ensure that the system is fully autonomous.

     

 

 

Automated Storage and Transfer Hub

To reduce human involvement in the parcel delivery process, an automated central hub aids with the storage and transfer of parcels which will be collected and delivered by the delivery robots.

The hub was designed with simplicity, cost, capacity, size, and efficiency in mind to allow for indefinite scaling to suit different business needs and space constraints.

The fabrication process:

Showing A360 Shared files

      

 

 

The Delivery Robot

A standardized parcel delivery attachment allows the robot to transport standard A5 parcels safely. The robot payload attachment is actuated using a servo motor to deposit the parcels at the hub.

The robot is also equipped with a 2D LiDAR and a depth camera to allow for autonomous navigation and accurate alignment to the hub.

The fabrication process:


      

rmf-navigation

RMF Navigation Graph

Each fleet of robots in RMF is provided a navigation graph superimposed on to the building's map, which constraints and allows RMF to generate navigation plans for the robots within the operational space.

 

 

Traffic Deconfliction

When multiple mobile robots are working in a given shared space, traffic conflicts and congestion are bound to happen if the robots are not properly managed. RMF performs traffic negotiations and deconfliction to prevent robots from congestion, traffic deadlocks and possible collisions.

Specialised RMF Features

     

web app user flow

Web App

Both the customers and vendors will be able to use the web application to make an order or view any existing order's status. They will need to sign up and login to their respective accounts, before being able to track any orders on the application. The interface focuses on simplicity and is straightforward to use, for users of different technological background.

The video on the left shows application flow from start to end.

   



Special thanks to the team at

 

TEAM MEMBERS

student Brendan Ong Yen Kie Engineering Product Development
student Chia Wing Yee Rachel Engineering Product Development
student Keane Hi Zhe Wei Engineering Product Development
student Lum Kai Wen Engineering Product Development
student Ling Zi Wei Rebecca Information Systems Technology and Design
student Yee Yi Xian Information Systems Technology and Design
border border