Drone Market ArduPilot Randy Mackay Japan Drones July 2017 - - PowerPoint PPT Presentation

drone market
SMART_READER_LITE
LIVE PREVIEW

Drone Market ArduPilot Randy Mackay Japan Drones July 2017 - - PowerPoint PPT Presentation

Drone Market ArduPilot Randy Mackay Japan Drones July 2017 version 0.1 Agenda 1.Intro 2.Drone market challenges 3.What is ArduPilot? 4.Current Projects 5.Working with Open Source 6.The Ultimate S&R system Drone market is getting


slide-1
SLIDE 1

Drone Market ArduPilot

Randy Mackay Japan Drones July 2017 version 0.1

slide-2
SLIDE 2

Agenda 1.Intro 2.Drone market challenges 3.What is ArduPilot? 4.Current Projects 5.Working with Open Source 6.The Ultimate S&R system

slide-3
SLIDE 3

Drone market is getting tough

Prices are falling dramatically

$1000

slide-4
SLIDE 4

Expectations are high

We are at the peak of the “Hype Curve” meaning expectations are very high

slide-5
SLIDE 5

Small start-ups are struggling

Three well funded kickstarter projects in trouble

2.2 Million USD (1 year late) 1.3 Million (16 months late) 3.5 Million USD (Failed)

slide-6
SLIDE 6

ArduPilot

ArduPilot is open source flight code:

Supports multicopters, planes, rovers, helicopters, antenna trackers

Partners program provides a forum for companies to collaborate on projects which are of shared interest

Other components:

Ground Stations (Mission Planner, Tower)

MAVLink, MAVProxy (communications)

DroneKit Software Development Kit

slide-7
SLIDE 7

Ardupilot Projects - Hardware

Pixhawk (ARM) NAVIO (Linux) Bebop2 QC SnapDragon PixRacer

slide-8
SLIDE 8

Ardupilot Projects - VTOL

Canberra UAV’s OctaQuad Plane for use in Outback Challenge

slide-9
SLIDE 9

Precision Landing

Thomas Stone of IRLock and SkySense inductive charging pad

slide-10
SLIDE 10

Multi-Vehicle (50 planes)

Advanced Robotics Systems Engineering Lab launches 50 vehicles in 27min

slide-11
SLIDE 11

Throw Mode

slide-12
SLIDE 12

Active Projects

Airspace Management

  • Agree protocols for no-fly-zone data
  • Final solution will integrate sensors like

uAvionix’s ADS-B receiver

“APSync” for Companion Computers

  • NVIDIA (and similar) companion computers

for vision based projects such as SLAM,

  • bject avoidance
  • Cloud integration

uAvionix’s ADS-B receiver Auvidea.eu’s Jetson TX1 carrier board

slide-13
SLIDE 13

Companion Computer (DroneKit) License: BSD Ardupilot License: GPLv3 Ground Station Wifi Flight Controller (Pixhawk2 – open source design) Sensors

Integrating with Open Source

Ardupilot is open source (GPLv3) but you can use a companion computer to add higher level features (even proprietary features) which allows your company to differentiate itself from its rivals. You get the reliability of the low-level flight code for free allowing you to instead invest in the higher level features. ArduPilot is at least as reliable as the leading closed system and you are not beholden to a particular manufacturer.

slide-14
SLIDE 14

Integrating example: Smart Shots

The Solo’s “Smart Shots” allows anyone to easily create very beautiful aerial videos by autonomously flying the drone in unique patterns while aiming the camera

  • appropriately. One of the most popular is “Orbit”. These are implemented in

Python using DroneKit which runs on the iMX6 companion computer.

slide-15
SLIDE 15

Actually making it work

Each company puts in at least one developer to work on that company’s priorities. Developers review each other’s work with senior developers making the final decisions. Weekly and ad-hoc meetings make the team function just like any other team. The way to make it work is to dedicate an English speaking developer to the team and get them to attend the weekly meetings and engage often with other team members. If that’s impossible, find a consulting company who can do that for you.

slide-16
SLIDE 16

Features of a great S&R system

Michael Day of Advanced Robotics Systems Engineering Lab launches 50 vehicles in 27min EnRoute’s UAV Van (includes Antenna Tracker)

High efficiency:

  • Multiple vehicles per operator
  • Relatively small drones
  • Fast deployment
slide-17
SLIDE 17

Highly Autonomous

Positions visible in real time Instructions to individual vehicles or groups of vehicles only when necessary

Operators can see live video or manually fly a single vehicle

EnRoute’s flight over SakuraJima May 2015

slide-18
SLIDE 18

Safe Real-Time Mapping

Avoids terrain and near-by manned vehicles Data sent back to Ground Station via Phone network or long range radio

slide-19
SLIDE 19

Data Distribution Locally & Globally

Raw picture, video and processed map data available locally via wifi and globally via cloud including sites like TomNod for crowd sourcing S&R

slide-20
SLIDE 20

More information

ArduPilot wiki: http://ardupilot.org/ ArduPilot software: https://github.com/diydrones/ardupilot DiyDrones: http://diydrones.com/ DroneKit: http://dronekit.io/ MAVLink: http://mavlink.org/ Developer email group: drones-discuss@googlegroups.com

Questions?