[ros-users] [Discourse.ros.org] [Aerial Vehicles] Autonomous Selfie Drone by Crazyflie using Deep Learning Models

Ildoo Kim ros.discourse at gmail.com
Thu Aug 10 07:21:05 UTC 2017


Our team has been working with Crazyflie platform to demonstrate how deep learning algorithms can be applied to robotics especially a nano aerial vehicles; thanks to Bitcraze development team~ :) To verify the reliability of our approach, our team has also applied the same approach to Petrone, made by Byrobot Co., one of the drone manufacturing companies in Korea

As rapid growth of deep learning algorithms has been made, we sincerely believe that autonomous drone tracking system using deep learning techniques will be generally designed and implemented in robotics community in short period of time. We hope that our project can be thought of as a first step towards the Robotics applied with AI

Here are some of the video clips of our demos; please enjoy it. 

1) Demo with Crazyflie
2) Demo with Petrone (on-going)

<img src='/uploads/ros/original/1X/d5a3349dbb64b7b59401093265c1a9bfa0662e31.jpg'>

**Hardware Specifications**

- Crazyflie 2.0: https://store.bitcraze.io/collections/kits/products/crazyflie-2-0
- 3g FPV Camera with transmitter, 5.8GHz 25mW 48CH: https://www.amazon.com/gp/product/B06XB6GDZF/ref=oh_aui_detailpage_o00_s00?ie=UTF8&psc=1
- FPV Camera receiver 5.8GHz 40ch RX for Computers: http://laserbgc.com/index.php?route=product/product&product_id=85
- 1g LiPo battery pack, 1s, 100mAh (for FPV camera)

**Software Specifications**

- Embedded Ubuntu 14.04
- ROS indigo ver. 
- Crazyflie ROS repo: https://github.com/whoenig/crazyflie_ros
- Petrone ROS repo (our version): https://github.com/ildoonet/PyByrobotPetrone
- Openpose ROS repo (our version): https://github.com/ildoonet/ros-openpose
  - *** Please refer to the README for installation ***
  - *** For more details about Openpose, please refer to the link below ***
  - https://github.com/CMU-Perceptual-Computing-Lab/openpose
- ROS video recorder (our version):  https://github.com/ildoonet/ros-video-recorder

**Descriptions of Key Features**

- Face Detection 
Using pre-trained data set using deep learning, it detects human face and calculates out the 3d vectors with unit of meter 
For the initial flight triggering, the drone must detects human face for 3 seconds.
Be careful if you are wearing a photo necklace, it will detect your photo and rotates motors :) 
- Human Body Detection
Using Openpose libraries from CMU (please refer to the link above), a drone is able to detect human body structures such as right/left shoulder, wrist, ankle, elbow, neck, etc
Using this technique, the drone can recognize human body and behave accordingly based on the defined flight motion 
- Special Flight Movements
When a drone detects certain patterns of human body, it delivers special flight motion listed below: 
  - Fade-away: when a person waves either left/right hand, 
  - Landing / Flip Motion: when a person raises both hands up 

If you have any comment or suggestion, please email us at curtis.v2 at kakaobrain.com. 
Thank you, 

P.S. Our team is eagerly looking for passionate drone engineers and AI researchers who would like to enjoy flying a smart-S drone, For more details regarding to the job description, please refer to the link below.

**Job Posting Link:**

[Visit Topic](https://discourse.ros.org/t/autonomous-selfie-drone-by-crazyflie-using-deep-learning-models/2414/1) or reply to this email to respond.

More information about the ros-users mailing list