Civilian Drone services: Concept, Strategies and Experimentation

Majed Alwateer

Bachelor of Science, from Canterbury university 2012, New Zealand
Master of Computer Science from La Trobe university 2014, Australia
Started my PhD at La Trobe university in 2015, Australia

Working progress of my PhD on civilian drone services is described below

Why Drone Services (Motivation)

In the coming years drones will be used in industries for different reasons. It will crowd designated regions of the sky and will be expanded with developed services. These services will be employed in many ways including: People who want their homes temporarily monitored while they are away for short and long duration (e.g., on vacation). To accompany someone from a designated starting point to a given destination (e.g., leaving work after hours).

A concept of technology- aided drone services is triggered in which users can use a mobile app to report when and where they feel unsafe. This will allow accumulation of data that can be used to assess the times and places (and situations) where people feel unsafe within the city, possibly useful for the relevant authorities and town planners.

The Concept Drone Services

Drone services can be defined as “Services involving the use of drones and digital technology to deliver valuable solutions to users in a commercial or non-commercial environment, often performing aerial physical actions to create better civilian environments”

Similar to cloud computing and other IT service architectures, drone services (i.e. drone computing) can be classified into three layers; infrastructure layer, platform layer and applications layer. These three models can accommodate most types of service that a drone could offer.

The components of the infrastructure layer consist of all technical infrastructures that are required to deliver a specific service such as the drone itself, the service provider, the third party owner of the drone, the drone station, the power station and possible payloads that can be mounted or used by an individual or multiple drones. All these components can be offered as IaaS.

The components of the platform layer comprise all software and frameworks that are placed on the infrastructure's components. For example, servers and their management, databases and their management, payload management (including mounted or remote), power management, communication management and decision management. Even though some of these components need to integrate to ensure a smooth end to end delivery, at certain degrees can be offered as PaaS.

As the application layer seen as the abstraction layer, it mainly aims at providing all required communication interfaces between all participants from the service provider to the end-users. This is including but not limited to the applications themselves, the certificates (e.g. security certificate) and web interfaces/dashboards. All these components can be offered as DaaS.

Drone Services (Proof of concept)

Despite the increase in drone popularity, still, there are numerous issues to be solved. But there are many beneficial applications that can be derived through the use of drones combined with other available, scalable and growing technologies such as mobile devices, smartwatches, and add-on electronic sensors. We propose and investigate an approach to using drones acting as servers (edge computing style) in order to collect data, provide Internet access, and process data for mobile users.

Data Collection

The act as a data mule where it collects files from client's devices while flying or hovering at a specific location.

Data Provisioning

The drone acts as a flying access point to provide internet access to clients via tethering.

Data Processing

The drone plays the role of a data processing partner that share its resources with other devices using the Honeybee technique

Crowd Powered Drone Services

As smartphones develop to be powerful and pervasive processing devices, their utilisation, likewise, keeps on expanding quickly. Yet, smartphone clients often encounter issues when running exhaustive applications because of resource deficiency and availability issues. Recently, at the same time, drones are rapidly finding their way into people's hands. As a result, increasingly, people are surrounded by devices with remarkable computational properties. We propose and examine two key concepts for utilising mobile devices, combining smartphones and drones as a crowd-powered resource cloud, in the context of providing smarter drone services, in particular, focusing on crowdsourcing for drone computations, and multi-drone service management using a new scripting language for coordinated flight paths of multiple drones. We describe our underlying model and experimentation with these concepts.

Crowd Powered Drone Network Architecture

Machine aided processing

The master controller distributed the task of processing to worker nodes (i.e., machine) using Honeybee.. Each worker mode that is engaged can process some photographs using the Android built-in face detection algorithm and send the results to the master

Human aided processing

The master controller distributed the task of processing to worker nodes (i.e., humans) using Honeybee. Each human that is engaged can process some tasks through a manual review (by answering ”yes” or ”no” to the following question corresponding to each image: ”Do you see a person in this image?”) and the results are sent to the master.

DroneScript is a scripting language that enables programmers to easily and quickly create missions for drones using commands issued from the master controller. The essential language elements are actions and missions.

Master UI

The master control user interface where the drone controller connect to surrounding drones and send missions

Proxy UI

The proxy user interface where the drone owner allow their drones to be discovered and hand control to the master control

Decision Making

We envision a future with companies providing civilian drone services to people - e.g., photo-taking on-demand for tourists from impossible (such as from off-the cliff) perspectives, object inspection, delivery, guarding, or helping someone check something out remotely. The broad aim of our work is to provide a framework for the cooperation between one or more stations and one or more drones as they are allocated to tasks.

Architecture: Decision Making For Drone Services Delivery

Things to consider in order to prepare for Decision Making:

A) Served area

It is important to know the nature of the served area to draw all of the related features,such as:

Maximum distance that a drone can travel.
The distance between a station and target
The distance between a drone and station
The distance between a drone and target

B) Station location

For a single station, with respect to the served area the station can be located at:

Edge of the served area
Middle of the served area

For multiple stations, the served area can be divided into many cells (what we call Drone Cells), each supported by two or more stations (top left, and bottom right, in the figure). Here, we focus on using one cell only with two stations.

C) Common factors

Drone speed

Battery capacity

Processing Time for each order

Range of communication,

Financial incentive

Environmental factors

Service type

Consumption & charging rate if applicable should be considered

Two-Layered decision model for task servicing by drones.

In order to study the decision process on drone services, we propose a two-layered task servicing a model that couples the "big picture" across-drones perspective of drone stations, and the dynamic local perspective of drones, in order to decide which task a drone should serve next, where tasks are first allocated to a drone's task set via an on-station strategy, and then drones select tasks to serve from their respective task sets via an on-drone decision-making strategy,