Blogs

Drones are now matured as a technology and are better understood by developers and users alike. We at FlytBase, have made efforts to provide you with some open source drone applications. These drone apps provide a base platform for your app development and can be customized as and when needed. We invite interested developers to contribute to this repository and help us grow this resource. So without further ado, here is a list of the 10 open source apps for drones you can use.

1. Joystick App (Download Mobile App, Source code)

This simple Mobile/Web App allows the user to control the drone using a joystick interface. You can take off/land your drone with the press of a button and fly it in any direction.

This app uses FlytBase Drone Navigation APIs to send velocity setpoints to your drone and eventually control Roll, Pitch and Yaw movements of the vehicle. You can find the source code and download the Joystick web app here.

2. Video Streaming App (Download app, Source code)

Live video streaming is a must have feature in almost any drone application.  This Drone App allows the user to view live video feed from the drone camera. The video is streamed is done over Wi-Fi from a camera attached to the Companion Computer mounted on your drone. You can find the source code and download the app here

3. GPS Follow Me App (Download app, Source code)

GPS Follow Me mobile app enables your drone to follow you around.

This is a sample mobile app using GPS-based positioning. When the follow-me mode is turned on, the GPS coordinates of the device (running the app) are used to send position commands to drone and it starts following the person carrying the device.  

Visit here for detailed information and source code.

Check out this video to see how GPS Follow Me App works:

4. AprilTag Detection App (Source code)

AprilTags are 2D barcodes developed for robotics applications and are now being used in various drone applications. This project integrates AprilTags detection with FlytOS. FlytOS takes input image from the camera and publishes processed image with detected AR tags. It can be used with any AprilTags family. You can find detailed information and source code here.

The video below shows a demo on detecting AprilTags with FlytOS:

5. Obstacle Detection using Sonar (Source code)

This Obstacle detection app uses Sonar connected to Arduino to detect obstacles from 6 directions and publish the distance data into ROS. The app also gives you the option to visualize live data received from the distance sensor.

You can write your own custom onboard application that consumes the sonar data and produces an obstacle avoidance behavior. Download the app and check out the source code here.

6. Visual Servoing (Source code)

This app uses gimbal APIs to control a 3-axis gimbal in order to keep the camera(mounted on gimbal) focused on an object of interest in the vicinity. FlytOS vision APIs are used to detect and track the object. Read the full blog on Drone API for Gimbal Control [How to do Visual Servoing]

Check this GitHub link for instructions and source code to setup this app for your drone.

Also, check out the video explaining how to do visual servoing using your companion computer:

Continue Reading

After a summer of misguided progress (apparently 3D printing was a terrible idea) and a busy school year, I decided to rebuild the plane and finish my project for my final high school project. Below are link to a condensed version of the presentation I gave, a time-lapse video of the build process and a flight video. Feel free to message me for details of the process or build.

Presentation:

https://docs.google.com/presentation/d/1qH7Sh3f_3Re3_vYbHrYAd1oMooiMFaM3Qlp4VSF-jtU/edit?usp=sharing

Flight Video:

https://drive.google.com/file/d/0B01p2a45VFilTU9ZV0o4cHk0OHc/view?usp=sharing

Build Video:

https://drive.google.com/file/d/0B01p2a45VFilSDNVOEdFTVJvSm8/view?usp=sharing 

Why are so many people looking forward to it and can Turnigy graphene lipo battery really can increase flight time for quadcopter? Today, I make a article about Turnigy graphene review for helping you to know what's the graphene battery and its performance.

What is the graphene lipo battery?

Graphene batteries is Newly developed a kind of new energy battery, which the use of lithium ion in graphene quickly a large number of shuttle between the surface and the electrode movement features.

There are at less 3 advantages of the graphene lipo battery as below:

• High capacity energy storage
• Shorten charging times
• Longer cycle
• Less weight
• More cheaper

Turnigy graphene review 1300mah

As above image you see, Typically I see the end temperature around 130F to 140F. Again, the Turnigy graphene pack performed much better than a standard LiPo. After more than 80% discharge at 27A it was barely warm at 108.5F! 

Conclusion

Can turnigy graphene increase flight time for quadcopter? Yes, but not exactly. In fact, according to our survey, graphene material just assisted as additives to increase the performance of lipo battery. So far, they are all more or less good for increasing the performance.

Gradually along with the mass production and large size problem, the industrialization application of graphene is speeding up, based on the existing research results, the commercial application of the first field would be on a mobile device, aerospace, new energy battery and drone field.

Now, there are many brands of graphene battery in the market such as Turnigy, Arris, Grazepony etc. Which one would you like to purchase? Please leave your messages on here, if you have any questions about the graphene battery.

20 amps ESC can be used for 2205 2300kv motors with 5 inches propellers. They are also compatible with motors that smaller than 2205 or slower than 2300kv. For example 2204 2300kv, 2206 1900kv (with 5 inch propellers).

If you are using 30amps ESCs, you should choose motors that are bigger than 2205 or faster than 2300kv, and propellers should be bigger than 5 inches. For examples 30 amps is required for the 2204 2700kv, 2204 3200kv, 2205 2700kv or 2206 1900kv(with 6 inch propellers).

35 Amp ESCs is required when you are using 6s lipo battery for your rc model.

and if you need 30 A ESC, you can view this review of Tattu ESC above

If you wanna buy it, here is Germany shop: http://www.gensace.de/tattu-esc-30a-blheli-s-2-5s-w-dshot-no-bec-prototype-1161.html

USA Shop: http://www.genstattu.com/ta-esc-30a-4pcs.html

Awesome post from Patrick Poirier in the OpenMV forums (OpenMV is my favorite computer vision board, and what I use on our DIY Robocars autonomous racers).

--------

This project is a variation of the original Rand'ys Red Balloon Finder implementation.
Based on this blog : http://diydrones.com/profiles/blogs/red-balloon-finder , I modified the Python scripts making now possible to test on any ArduPilot based Quadcopter on a low budget , relatively easy to implement controller.

Code: Select all

     > #This is the configuration script, allowing usage the configuration file       > import balloon_config      > #We need these modules (installed with dronekit ) to control vehicle>     > from pymavlink import mavutil      > from dronekit import connect, VehicleMode, LocationGlobal        > # connect to vehicle with dronekit      > #MAIN      >             # only process images once home has been initialised      >             if self.check_home():       >                 # check if we are controlling the vehicle      >                 self.check_status()       >                 # look for balloon in image      >                 self.analyze_image()       >                 # search or move towards balloon      >                 if self.search_state > 0:      >                     # search for balloon      >                     self.search_for_balloon()      >                 else:      >                     # move towards balloon      >              self.move_to_balloon()# move towards balloon      > # move_to_balloon - velocity controller to drive vehicle to balloon      > # calculate change in yaw since we began the search      > # get speed towards balloon based on balloon distance      > # apply min and max speed limit      > # apply acceleration limit      > # calculate yaw correction and final yaw movement      > # calculate pitch correction and final pitch movemen      > # calculate velocity vector we wish to move in      > # send velocity vector to flight controller      >       send_nav_velocity(pitch_final, yaw_final, speed)       >     # complete - balloon strategy has somehow completed so return control to the autopilot                >      # stop the vehicle and give up control      >        # if in GUIDED mode switch back to LOITER  

OpenMV Script
The Red Balloon Finder is a typical colored BLOB Detector/Tracker. 
We are adding a serial output to PORT 3 so the x-y location and blob width & height can be transmitted to the RPI Zero.
uart_baudrate = 9600
uart = pyb.UART(3, uart_baudrate, timeout_char = 1000)
uart.write("%d ; %d ; %d ; %d \r\n " % (blob.cx(), blob.cy(),blob.w(),blob.h()))



Some theory
In vision based systems, there are many types of hardware/software configuration tailored for sp[/img]ecific applications: Visual Servoing, Visual Odometry and Visual Simultaneous Localization And Mapping (SLAM). In this project we are using the former type of system: Visual Servoing that is designed to:
• Take off and Landing
• Obstacle Avoidance/Tracking
• Position and Attitude control
• Stabilization over a target

The main idea of Visual Servoing is to regulate the pose {Cξ,T } (position and orientation) of a robotic
platform relative to a target, using a set of visual features {f } extracted from the sensors.


Source: Survey on Computer Vision for UAVs: Current Developments and Trends 

Randy's Target Tracking is an Image Based Visual Servoing (IVBS), where the 2D image features are used for the calculation and control values. We exploit a hybrid method where the size of the object is known -a priori- making the estimation of distance along the Z axis possible. In the example below, were the system is following a moving target at a fixed distance, we can relate the target position to the camera projected plane


Source: 3D Object following based on visual information for Unmanned Aerial Vehicles

In this Tracker, we apply a color and shape (blob) filtering in order to extract a location on the camera plane.
We need to extract position and and estimate the distance by computing the knowned object size with the camera parameters: 
lens_mm = 2.8
lens_to_camera_mm = 22
sensor_w_mm = 3.984
sensor_h_mm = 2.952
x_res = 320
y_res = 240

The field of view derives both from the focal length of the lens and the size of the image sensor.
h_fov = 2 * math.atan((sensor_w_mm / 2) / lens_mm)
v_fov = 2 * math.atan((sensor_h_mm / 2) / lens_mm)
In degrees : h_fov = 70.85 , v_fov 55.60


http://chrisjones.id.au/FOV/fovtext.htm


The values coresponding to the object positions and size are transmitted from OpenMV to RPI Zero using serial communication. The python scripting module OpenMV.py reads these values and pass them the Analyze_image() process within Balloon_Strategy_2.py in order to transform position and distance. Here is how we compute the distance:

Code: Select all

    # get distance     self.balloon_distance = get_distance_from_pixels(size, balloon_finder.balloon_radius_expected)        ====FROM find_balloon.py====      # define expected balloon radius in meters      self.balloon_radius_expected = balloon_config.config.get_float('balloon','radius_m',0.3)     # get_distance_from_pixels - returns distance to balloon in meters given number of pixels in image      #    size_in_pizels : diameter or radius of the object on the image (in pixels)     #    actual_size : diameter or radius of the object in meters (0.3)     def get_distance_from_pixels(size_in_pixels, actual_size):     # avoid divide by zero by returning 9999.9 meters for zero sized object      if (size_in_pixels == 0):         return 9999.9     # convert num_pixels to angular size     return actual_size / balloon_video.pixels_to_angle_x(size_in_pixels)           ====FROM Balloon_video.py====     # get image resolution     self.img_width = balloon_config.config.get_integer('camera','width', 320)     # define field of view     self.cam_hfov = balloon_config.config.get_float('camera','horizontal-fov', 70.85 )     # pixels_to_angle_x - converts a number of pixels into an angle in radians      def pixels_to_angle_x(self, num_pixels):         return num_pixels * math.radians(self.cam_hfov) / self.img_width

Building the Tracker System
Depending on the type and size of UAV you are planning to use, the implementation may vary. This build is for a Quadcopter 330 mm size, with a pixracer Flight Controler and a single 5Volts 3 Amps BEC. It is assumed that the UAV is completely functionnal, tested and tuned to be ''rock solid '' in Loiter & Guided mode.





Loading the code, setting parameters & testing 
Test&Development environment:
⦁ RPI Zero to a powered USB hub, keyboard, mouse, WIFI, FTDI serial and HDMI monitor
⦁ Window based PC for OpenMV & Mission Planner
⦁ Linux (or VM) based PC for SITL


OpenMV script
Open IDE
Load the Script
Adjust the color filter using Tools/Machine Vision/ Threshold editor
Make sure you are using LAB color space
Cut the Values & Paste them in the OpenMV script. These values must be inserted in the appropriate thresholds filter code (generic red thresholds for our example):

Run the test and confirm that it is tracking steadily
When satisfied, Tools/Save Open script to OpenMV 

Python Script
On your RPI, open a terminal window and clone the program
git clone https://github.com/patrickpoirier51/OpenTracker

_The code is still a WIP , sorry for the mess _

Move to the OpenTracker/script and with an editor (simply open the file with the file manager), adjust the parameters of ballloon_finder.cnf , if required. You can test if the OpenMV to RPI Zero connection is working correctly by launching OpenMV.py using command line : sudo python openmv.py or using IDLE2 editor and running script. Remove the # before print command in order to see the values on console. You can run commands without sudo but it might happen sometimes that you dont have all the priviledge to access codes and devices; its up to each users to test with or without sudo.

Hint: 
Ckeck for activity light on FTDI (Show / Hide object) == Knowing the pattern will be helpfull once you are testing live

Once satisfied, comment out the print command, save and you are ready for test.


Testing in SITL
Practice makes good and using SITL may save your day 
Leaving the
You can connect the RPI Zero Ftdi Usb to Serial converter to a second FTDI USB to serial on a Linux based computer (dont forget to cross the XMIT to RX(
Launch SITL:
/home/Ardupilot\ArduCopter/$ sim_vehicle.py --console --map --aircraft=balloon --out udp:missopn.planner.pc.adress:14550 --out /dev/ttyUSB0,115200

On the RPI start the Tracker on within a terminal session with sudo python balloon_strategy.py or using IDLE2 editor and running script. Make shure that we connect using Mavlink Serial- USB
connect ( '/dev/ttyUSB0' , baud=115200)

You should see on the RPI console the connection to vehicle, the parameters initialisation sequence and the tracker waiting for command.

On SITL you can initiate this sequence:
⦁ mode loiter
⦁ arm Throttle
⦁ rc 3 1800
⦁ takeoff 30 (wait until it reach this altitude)
⦁ rc 3 1500 (keep altitude in LOITER)
⦁ mode guided

Once in guided, the RPI takes command and you will see the quadcopter start turning, serching for the object.
You can test by ''showing'' the object to the camera for a brief period of time, and you should see the " "Found Balloon at heading:%f Alt:%f Dist:%f" " message appearing. Hide the object and wait until the message "Balloon was near expected heading, finalizing yaw to %f" message appears , starting the final alignment to object , and then you can ''show'' the object and start chasing when you see "Finalized yaw to %f, beginning run"

Now, the simulated vehicle will follow the object on the camera, Looking at the OpenMV IDE you will be able to judge if the simulator react accordingly to the target position on the camera screen. Basically the vehicle will move left-right and up-down (altitude) according to where the object is located in reference to the center of the camera, if its dead center, the simulated quadcopter should go straight with no altitude change.

If you hide the object:

Code: Select all

     # if we lose sight of a balloon for this many seconds we will consider it lost and give up on the search      self.lost_sight_timeout = 3      the tracker will give up : "Lost Balloon, Giving up"     and will go through this sequence:     # complete - balloon strategy has somehow completed so return control to the autopilot               # stop the vehicle and give up control     # if in GUIDED mode switch back to LOITER

Once in LOITER you can run another test just by switching back to GUIDED.




Quadcopter Integration
Here is a picture showing the typical installation:


ArduPilot
Adjust the parameter of the Serial (UART - SPEED - MODE) corresponding to the actual speed (115200) and for MavLink protocol.

Configure the flight modes so its easy for you to switch from LOITER to Guided on the Transmitter.


Headless configuration on the RPI Zero
In order to get maximum speed and power, disable the desktop mode (configuration menu or raspi-config)
There are many ways to get the script running automatically, the simplest being launching it within /etc/rc.local
edit file sudo nano /etc/rc.local, and add this:
python /home/pi/OpenTracker/blob/scripts/balloon_strategy_2.py &

"&" is used to make run python in background
Save and reboot 
Check for activity light on FTDI (Show / Hide object) 



Live Testing
Choose best : Field/Wind/Brightness/Mid Day

Choose the appropriate Object for Target == Red Balloon 
Place Target at least 2 Meter from ground 
Final adjustment of Target with OpenMV IDE using Threshold utility
Set the best Color Filter Mask and Blob Size
Scan around and Make sure you dont get "False Positive" 

Start Quad copter
Check for activity light on FTDI (Show / Hide Target)

Arm
TakeOff Loiter
Switch to Guided (Ready to switch back)

Check Sequence, Analyse Results, Repeat.... Have Fun !!!

Dear ladies and gentlemen,
I present a plan of Unmanned Underwater Vehicles in the future.

The biomechanical simulation of crawling swimmers (above) 

Other examples

The biomechanical simulation of hand movement :
(It is similar to jellyfish propulsion mechanism.)

The biomechanical simulation of rays' fins :

The biomechanical simulation of anomalocaris :

These creations were born from a discovery of the theorem of voluntary movement of natural animal as well as an invention of the universal mechanical linkage.

--

My work in progress :

In May, I was invited to ARTIS the Royal Zoo in Amsterdam. The director request me for the physical implementation of artificial animal. So, I'm proposing to rapid prototype using 3D printing and other affordable hardware and software.

--

Industrial trends :

NVIDIA Isaac (Unreal Engine 4 based virtual simulator) can train AI of such as robot that runs with wheels, robot that fly in the sky with propellers, robot that moves underwater with screws. So I can produce mechanical legs in place of wheel, mechanical wings in place of propeller, and mechanical fins in place of screw. The burden of mechanism analysis (work to obtain the joint parameters to optimal exercise) which occupy the majority of my production process will be greatly improved by Isaac and AI.

With the new image guided application platforms based on Augmented Reality such as the Microsoft HoloLens, I'm considering to contribute to 3D drone application and medical treatment / rehabilitation / scientific research.

--

Background :

I'm a Romandy(French speaking Swiss)-based independent scientist.

In November 2005, when I was watching the demonstration of androids made in the 19th century at the museum of Neuchatel, I met a NAS engineer (NASA Advanced Supercomputing division).
He was invited by NASA after developed many cars in Subaru.
He was a super generalist who understood all the manufacturing processes during this period. Therefore, NAS needed his outstanding overall wisdom for the high end CAD design and manufacturing of the Mars probe.
He instructed me with a lot of advice. He immediately noticed my potential talent through his experiences and was expecting me. He predicted that I would become famous little by little.
Today, open source software has evolved, and hardware has become cheaper and convenient.
So, I made the effort to be a generalist like him.

In 2006, 2007 I settled in Sainte-Croix. then participated in the project of François Junod (He is a great master of automata creator.) and Leganess city of Spain, and actually supervised the production of large artificial horse.

Later I moved to La Chaux-de-Fonds. I learned the game theory and the principle concept of biomechanics of Chinese martial arts.

In 2012, under the approval of the French state, I launched a new project. And I made a revolutionary invention. At that time, I found a new theory of voluntary movement which was submitted by Abraham-Louis Breguet, a physicist and horologist in his book "Essay about the animal force and the principle of voluntary movement" in the early 19th century. I decided to promote a new research and development by combining his theory in my inxentions. This is my philosophical approach. then I was invited to the Breguet Museum at Paris.

I study the wisdom of Huguenot and anti-Qing Dynasty resistance forces. It is the same as during the Renaissance period, Leonardo da Vinci studied the wisdom of ancient Greece and Roman culture from Arab and Bizants. Their internal reform efforts are an open system that has external interactions. That become a philosophy lead a natural science which find the trust in providence to evolve the human race and that future.

醫生為機匠 The Doctor as Mechanic

Guys, I thought you might appreciate our new drone SUWAVE (Sherbrooke University Water-Air VEhicle). It is a new seaplane concept that should allow us to hop from lake to lake using solar energy. It uses new VTOL takeoff and landing techniques, while being robust to waves and wind during standby on a lake. Such platform could use lake as recharge station to monitor a certain area or to travel extended distance. More info on IEEE Spectrum. 

Alexis 

There are massive amount of quadcopter brushless motor brands in the market, What factors should you consider when choosing a given quadcopter motor brand? and Which one is the most popular one? Below are listed most brushless motor brands for you.

Cobra

Cobra brushless motors are designed to offer performance and value for your quadcopter. Built from some high-quality components, they provide top-level performance for years. Featuring an integrated motor shaft, they ensure that prop adapters do not spin off during operation.

FPV model

These are high quality and affordable brushless motors widely used for quadcopters as well as other multirotors. Offering both CW and CCW frame options, they integrate high power N42 magnets and CNC machining.

DYS

With a convenient weight, this is a brushless motor brand that offers some of the greatest motors for your quadcopter and other multirotors in the market. With an awesome shaft and exemplary KV rating, perhaps you need to try it out.

Multistar

This is a high-efficiency brushless motor which offers the best spinning power. Given its weight coupled with its awesome shaft and extra long leads, this is a great quadcopter motor.

LD power.

Designed for FPV drones, this high power and high quality quadcopter motor is engineered with high quality magnets and bearings for an outstanding performance. Brushless in nature, they are not only high quality but also come in a pocket-friendly price. With CNC motor cases, they are quadcopter motors worth trying oiut.

EMAX

Given their price coupled with a 3S LIPO battery and a diameter of 23mm, this is a quadcopter motor whose performance is unparalelled. With an awesome shaft of 2mm and an awesome weight, this is a product you can trust.

Sunnysky

Rapidly gaining a reputation in the quadcopter world, this motor is well designed for your quadcopter as well as multirotor. Brushless in nature, they come with prop spinner and mounts for exemplary performance. With an ability to generate a high thrust of up to 2kg, you can trust them with your large quadcopter.

Xnova

Xnova is a brand that has been refreshing its wide range of quadcopter motors day in day out. Providing high performance and a quality which is second to none, you can try these quadcopter motors for impressive results.

Edge

With an impressive 2300KV and an awesome stator length, this is a quadcopter motor worth trying out. At a pocket-friendly price, this is a motor for anyone looking for a quality product without having to dig deeper into their pockets.

Gemfan

To use these high quality quadcopter motors, all you need to do is use the correct propeller sizes since using wrong propeller size could affect their performance. However, they are an amazing quadcopter motor brand worth your cash.

T-motor

Providing an impressive performance and a longer flight time, T-motors are the most ideal motors for your quadcopter if you want your quadcopter to stay in the air for a long time. Designed to ensure the least temperature increase which translates to less heating up, they are quite reliable.

Tornado T1

Tornado T1 motor is simply fascinating. With an aweome magnet which provides an impressive electromagnetic power, you can trust them with your quadcopter. In fact Tornado T1 motors are well built for awesome performance.

Lumenier RX

Designed to give top of the line performance for all your RC models, these quadcopters are not just unique but exemplary. They are engineered from ground up to offer reliable power and enough spinning ability for your quadcopter. Given their strong stainless steel shaft, they are what you need.

Storm

Storm quadcopter motors are designed to offer the best prop spinning for your quadcopter. And yes, designed to conveniently work with props of different sizes, you can always have them for a quadcopter of any size.

RCX

RCX are brushless quadcopter motors which are designed for smaller multirotors and quadcopters. Small yet powerful, they have a capability of running your quadcopter for a considerable amount of time.

Eachine

Eachine quadcopter motors offer an awesome flight time as well as an amazing performance. In fact, their performance is top-notch and have been able to earn a reputation with most quadcopter manufacturers.

Racerstar

Racerstar are the perfect quadcopter motors for your fast moving quadcopter., With an amazing shaft, their KV rating is exemplary. And yes, the price is pocket-friendly. Try them today for a chance to enjoy amazing performance.

Hobbywing

With a long term reputation, hobbywing are quadcopter motors whose track record speaks volunes. Having provided best results for along time, you can trust them as well.

You need to consider below parameters when choosing your quadcopter brushless motor:

1. Voltage.

2. Thrust.

3. Efficiency.

4. Current draw.

5. Response.

6. Props.

7. Vibration.

8. Temperature.

9. Type of magnet.

10. The shaft.

Therefore, What's the best brushless motor for quadcopter? Brushless motors are typically 85–90% efficient or more, higher efficiencies for a brushless electric motor, of up to 96.5%, whereas DC motors with brushgear are typically 75–80% efficient.

Brushless motors are great for several reasons:

They’re incredibly efficient so they generate very little heat,no wear or contact items (such as brushes) and can spin incredibly fast.

Power density is incredibly important because on anything that’s flying you want to do everything you can to minimize weight.

Nobody complains when components are inexpensive.

Looking like a device straight out of science-fiction, drones have taken the world by storm. Also called Unmanned Aerial Vehicles, drones are being used by private industry and government. Whether being used to provide aerial views of an archaeological dig or to provide stunning shots for a blockbuster motion picture, there's no doubt drones are here to stay. As a result, they are changing the tech industry in ways no one ever imagined just a few years ago. 

A Multi-Billion Dollar Market

According to the Teal Group Corporation, which specializes in defense and aerospace research, almost $90 billion will be spent on drones over the next 10 years. A staggering figure, it demonstrates the almost unlimited potential of these devices. In fact, the tech industry is in fast-forward mode, with various prototypes being developed and tested not only in the United States, but also Australia, Switzerland, and many other nations.

Open-Source Technology

Along with the various shapes and sizes drones can come in today, the tech industry is rapidly developing new types of hardware and software that are being used for numerous applications. Known as open-source technology, these simplifications to the equipment and IT processes that help pilot the drones have led to great reductions in cost, enabling hobbyists as well as researchers to use drones for more and more applications.

Replacing the Mail Truck

If you've got an IT degree and want to work on something that will impact society, then drones are your ticket. For example, Amazon.com is currently partnering with many Silicon Valley companies to develop drones that will deliver packages to a person's front door. Current prototypes, which can carry packages of five pounds or less, could one day deliver to as many as 86 percent of Amazon's customers. 

The Hacking Threat

 As with most of today's technology, the threat of being hacked is always looming. This is especially true with drones, and as a result tech industry companies are investing millions of dollars to design software that will be virtually hack-proof. However, hackers have already released design and software specs that allow drones to be taken over by other individuals, creating what are known as "zombie drones." Because of this, tech companies are making cyber-security a top priority. 

With drones continuing to grow in popularity, there appears to be no end in sight for their uses. Therefore, people possessing IT degrees will be in high demand to meet the constant challenges of this ever-changing technology.

This video shows the Maiden flight of the Arduino-compatible STM32L4 "Ladybug" brushed-motor  flight controller from Pesky Products (enabled by the Arduino STM32L4 package from Grumpy Old Pizza), running my Hackflight firmware (which already runs on a variety of flight controllers, as well as a simulator).  I used a 3DFly frame, Lemon RX satellite receiver, Walkera Devo7E transmitter running DeviationTX firmware, and standard 8.5mm brushed motors for the Hubsan X4 and similar vehicles.  

The goal is to have a very inexpensive, state-of-the-art flight controller for nano-scale quadcopters, with a minimalist, object-oriented firmware design (around 1500 lines of C++) that you can program with the Arduino IDE, enabling all sorts of cool projects. The board has pins for I^2C, UART, and GPIO, allowing it to be connected to the new ultra-lightweight, low-power Arduino-ready sensors (like this one).

Instead of a traditional IMU, the board uses an EM7180 SENtral Sensor Fusion coprocessor, which performs much of the work traditionally done by the flight-control firmware. The EM7180 also incorporates a barometer, for altitude estimation. The main processor is an STM32L4, a low-power-consumption ARM Cortex M4 running at 80 MHz, with a Floating-Point-Unit.

The next iteration of this board (currently being built) will feature a standard Spektrum DSM connector for the receiver (which is currently plugged into the RX2 port on the board for convenience until the new iteration arrives).  It should also be possible to use a CPPM receiver on the same connection, with a slight modification to the firmware.  Once the prototype has been tested, I will be looking to produce the assembled boards in bulk, for use by students, researchers, and hobbyists.  

The major constraint for effective use of UAS in agriculture is the time and complexity of data processing. With current solutions, it can take hours of processing or data upload just to visualize even a small area of a farm, research plot, or golf course. We at intelliSky sought to find a solution. intelliSky is proud to introduce intelliSky Studio. intelliSky Studio focuses on single capture analytics, allowing you to capture only what you need, reducing the data set requirements 7 fold or more, and, analyze the data in a matter of seconds. By understanding the specific area’s you want to focus on and only gathering the data you need, multispectral data is now a practical solution for rapid analytics on your farm, golf course, or research plots!

Our software does multi-band registration using sub-pixel alignment algorithms, dark current corrections, vignetting corrections, radiometric calibration, distortion corrections, and custom visualizations.

Some of the primary benefits include:

• Rapid Processing (10 seconds or less on most computers)
• No overlap required
• Capture only what you need
• Daily analytics now possible
• Custom index generation
• Custom colorizations
• Exportable images and statistics
• Polygon tool for statistical and historical analytics
• Great complementary tool to traditional mosaicing methods

At the moment it is designed for the Parrot Sequoia™  but plan to add support for additional sensors and analytic tools very soon!   

Sign up for a free trial today at www.intelli-sky.com

Dear community members,

During this ICRA we presented our work on "Uncertainty-aware Receding Horizon Exploration and Mapping using Aerial Robots".

This work is now accompanied with an open sourced ROS toolbox available at: https://github.com/unr-arl/rhem_planner 

These videos present the planner in operation:

In case you find it interesting, we would be happy to support you in order to fully integrate it in your research.

Best,
Kostas

http://www.autonomousrobotslab.com

Hi, guys,

Recently we were conducting pyrotechnics tests on drones for the purpose of using them drone shows.

We shot some video footage of the pyrotechnics tests and thought you guys might have an interest in seeing some drones shoot fireworks. Using UgCS DDC (Drone Dance Controller) it would be possible to control multiple drones equipped with light effects or other show effects in a synchronised manner.

Here's the video:


Get the newest version of UgCS here

Safe flights,
UgCS Team

Have a look at the wearable groundstation, that I have built as cockpit for my Boeing 787-like drones avionics suite FlightZoomer...

The wearable groundstation has these features:

• A solid carrying bracket allows to put on the whole groundstation without neckstrap
• Perfect symbiosis between man and machine
• You can sit, walk or drive with the groundstation put on
• The tablet screen sits 30cm/12" in front of the eyes, which gives a very realistic feeling of sitting in a real cockpit
• Like you step into a real cockpit, you can put on this groundstation
• The RC transmitter is a Graupner MX16
• All the utilities to handle the FPV video signal are built in (RC832 receiver, UVC USB capture, battery)
• Weights less than 6kg (= you hardly notice the weight when put on)
• Formidable sunshade solution, the gap between eyes and screen is almost entirely covered, the angle of the sunshade can be adjusted
• There is a Microsoft Surface dial to control all relevant modes of the autopilot using a single turning knob (see my last blog post here).
• The dial can be transport locked, using a spring loaded bracket

Gone are the days where drones were used only for taking pictures and videos. Nowadays, you can perform a variety of meaningful tasks using your drone. Drones are starting to automate tasks in many industries like agriculture, construction, photography, disaster management and real estate. Knowing how to build your own custom application, goes a long way in helping you achieve these tasks

In this post, I will tell you how you can develop your own mobile app for your Drone. I will be using FlytOS, a software framework provided by FlytBase. FlytOS is an ideal platform to develop research and commercial drone applications.

Step 1: Setup Your FlytBase Account and Download FlytSDK

Step 3: Download and Install FlytSIM and Test Your App

FlytSIM is a ROS-Gazebo environment where you get a simulated version of your drone application. It is a great tool to test your drone app before the first flight. FlytSIM gives you an idea how your drone will behave in the real world.

You can find the step-by-step guide to install FlytSIM here.

Every app you develop using FlytSDK will require you to enter the IP address of the device running FlytSIM(usually your laptop).

After installing FlytSIM, run it on your system and connect it to your Wi-Fi network. Connect your phone to the same Wi-Fi network and launch your app.

Now, you can start testing your application in the simulation environment.

While testing, be sure to check all the functionalities as it is a great way to find out bugs/errors if any.

Step 4: Install FlytOS on Your Companion Computer for Your Drone

Here is a step-by-step guide to install FlytOS . It supports multiple companion computers and provides dedicated support. Choose the companion computer you are using from the list, and download FlytOS. Once the installation is complete, you must activate the license for the navigation APIs to work. It's free to use.

Now, you are ready for your first flight.

Step 5: Get Ready for Field Test

Once the FlytOS is setup and your drone is ready, you can test your custom app. If you have followed the steps correctly, everything should work as expected.

Before you fly, be sure to check out the safety guidelines.

Congratulations you have successfully developed and deployed your own mobile app for your drone.

You can refer FlytAPI for further development and assistance.

Happy Flying!

At KU Leuven we have been developing a new generation of drones to double the speed and range of conventional multi-rotor drones.
For quite some years already, multi-rotor drones are the preferred type of drone for typically applications such as filming or inspection, in which the drone flies at low speeds and usually stays rather close to its pilot. On the one hand, the design of these drones is not very well suited for covering long distances at high speeds, which are beneficial in multiple applications such as parcel deliveries or emergency services. On the other hand, fast and efficient fixed-wing drones require a person or catapult to take off and a runway, parachute or net to land which limits their possibilities in fully automated missions.

Our team has been focusing on improving the speed and range of multi-rotor drones. We therefore combined the benefits of fixed-wing drones with those of multi-rotor drones resulting in a hybrid solution which performs a transition from hover flight into an efficient cruise flight. The only moving parts of the drone are the four propellers which are used to provide lift, thrust and control. This results in a robust, lightweight and low-maintenance solution.  Back in 2014, our first prototypes took off.  These designs made a transition from hover to cruise flight of 90°, with a large wing producing 100% of its required lift in cruise flight. Flying in windy conditions, however, was challenging due to this large wing. Therefore, we decided to improve the wind resistance by reducing the wing size and compensate the loss in lift by improving the aerodynamics and only make a partial transition such that the propellers still produce part of the lift in cruise flight. The result is a very wind tolerant, efficient and fast new generation of drones… The CargoCopter.

Our designs have been evolving fast: a parametric selection algorithm optimizes every new CargoCopter design for its mission-specific requirements. The designs are specifically engineered to be suited for lightweight 3D printing. Thanks to this agile methodology, we were able to design, manufacture and test-fly dozens of prototypes with gradual improvements leading up to doubling speed and range compared to standard multi-rotor drones.

The CargoCopters designs are meant to transport payloads up to 5kg, achieve speeds up to 150km/h and cover distances up to 60km.

This week the Dronecode/PX4 team will release its biggest and best update, version 1.6. You can see a glimpse of the testing that went into it with our public test log server here. 

Along with all the automated code and flight simulator testing, there are more than 100 hours of real-world flight testing (above and beyond many times that many hours by the beta testers) by the full-time Dronecode Test Team, as show here.

Excerpted from a new post at our DIY Robocars sister site: 

It’s now possible to buy small scanning 2D Lidars for less than $400 these days, which is pretty amazing, since they were as much as $30,000 a few years ago. But how good are they for small autonomous vehicles?

I put two to the test: the RP Lidar A2 (left above) and the Scanse Sweep (right). The RP Lidar A2 is the second lidar from Slamtec, a Chinese company with a good track record. Sweep is the first lidar from Scanse, a US company, and was a Kickstarter project based on the Lidar-Lite 3 1D laser range finder unit that was also a Kickstarter project a few years ago (I was an adviser for that) and is now part of Garmin.

The good news is that both work. But in practice, the difference between them become very stark, with the biggest being the four times higher resolution of the RP Lidar A2 (4,000 points per second, versus Sweep’s 1,000), which makes it actually useful outdoors in a way that Sweep is not. Read on for the details.

First, here are the basic spec comparisons:

Bottom line: RP Lidar A2 is smaller, much higher resolution, and better range indoors (it’s notable that the real-world RP Lidar performance was above the stated specs, while the Scanse performance was below its stated specs). The Scanse desktop visualization software is better, with lots of cool options such as line detection and point grouping, but in practice you won’t use it since you’ll just be reading the data via Python in your own code. Sadly the Scanse code that does those cool things does not appear to be exposed as libraries or APIs that you can use yourself.

In short, I recommend the RP Lidar A2.

I tested them both in small autonomous cars, as shown below (RP Lidar at left)

Both have desktop apps that allow you to visualize the data. Here’s a video of the the two head-to-head scanning the same room (RP Lidar is the window on the right)

You can see difference in resolution pretty clearly in that video: the RP Lidar just has four times as many points, and thus four times higher angular resolution. That means it can not only see smaller objects at a distance, but the objects it does see have four times as many data points, making it much easier to differentiate them from background noise.

As far as using them with our RaspberryPi autonomous car software, it’s a pretty straightforward process of plugging them into the RaspberryPi via the USB port (the RP Lidar should be powered separately, see the notes below) and reading the data with Python.  My code for doing this is in my Github repository here.  We haven’t decided how best to integrate this data with our computer vision and neural network code, but we’re working on that now — watch this space

Read the rest here

Hi! Version 1.22 was just released of SidePilot, the iOS Groundstation app for Arducopter (Inlcuding 3DR Solo) and PX4

New features include:

• Save video to camera roll (Using 3DR Solo)
• Better Fullscreen video
• Dedicated Shots button
• Offline Waypoints
• Timestamp for dataflash logs
• More waypoint options (Camera & Servo trigger)
• Shots! Our version of smart shots, now available to all Arducopter drones!
  • Follow me (Both static and dynamic altitudes)
  • Cable Cam (Define two points and automatically transition between then)
  • Orbit (Circle a POI at any distance to create fantastic cinematic shots)
  • More coming!

Other Features:

• 3DR Solo live video feed & GoPro control
• GPS Location of drone
• Set, edit and delete waypoints
• Arm/Disarm
• Display VFR information such as altitude, yaw, pitch and roll.
• Voice alerts for mode changes and other important information
• Edit Parameters
• Log flights for future reference
• GPS Status of drone
• Distance from home/device
• Battery voltage display and warnings
• Calibration of battery monitor and accelerometer
• Light and Dark themes
• Manage multiple BLE Repeaters
• Live video feed from drone (if supported)
• Change flight modes
• And many more

For more info please visit http://sidepilot.net

Download on the App Store Here

You must have either the BLE Repeater device that accompanies the app, or a Wifi connection to your drone (such as 3DR Solo). You can purchase the repeater from here: http://sidepilot.net/store

If you purchase the BLE Repeater, the app is included for free!

If you have any questions please ask!

An Australian IT services company combined an Amazon Alexa and a service called "what3words" that translates GPS addresses into three unique words to create a MAVLink-compatible drone that can be commanded entirely by voice. The video can't be embedded, so see it here.