TonyleeKorea Project Documentation

site overview

Welcome

To the most advanced hardware and software ecosystem for enterprise drones.

Sky-Drones Online Documentation

We provide the entire full-stack hardware and software ecosystem for enterprise drones. This online documentation covers the how-to, what-if, and just-in-case elements of our whole product range, including:

SmartAP Airlink: The most advanced AI Drone Avionics: autopilot, mission computer and LTE module.

smartLink: Drone connectivity system with integrated onboard computer. Provides direct datalink, LTE connectivity, video streaming, remote control from the Cloud and more.

skydrone-cloud: Cloud-based web application for drone fleet management, asset management, performance monitoring, mission planning, real-time control and post-flight AI analytics.

smartAP-gcs: Cross-platform ground control station software for vehicle configuration, mission planning and flight control.

smartAP-pdb: Power transfer from the battery to ESCs / motors and power supply for the

smartAP-max: Flight Control System aimed at a wide range of purposes for consumer and professional applications with a compact main core and externally connected peripherals.

smartAP-Pro: Industry-leading Flight Control System with outstanding reliability, precision positioning, and control accuracy for professional applications with various peripherals already integrated onboard.

Hardware Products

SmartAP AIRLink - The most advanced AI Drone Avionics

SmartAP AIRLink Enterprise Edition airlink stands for Artificial Intelligence & Remote Link. The unit includes cutting-edge drone autopilot, AI mission computer and LTE connectivity unit. Start your enterprise drone operations with and reduce the time to market from years and months down to weeks.

SmartLink - Drone Datalink & Onboard Mission Computer

Broadband Digital Datalink with Integrated Onboard Computer and LTE connectivity. Two HD video channels, telemetry and control with ultra-low latency and a range of up to 20 km.

SmartAP MAX and SmartAP PRO - Drone Autopilots

Legacy Sky-Drones flight controllers - SmartAP PRO and SmartAP MAX

SmartAP GNSS, SmartAP RTK, SmartAP PDB Hardware can be purchased in the official .

Software Products

SmartAP GCS

Configure, plan missions, and operate multiple drones with our cross platform ground control station software. The LTE connectivity allows users to sync their data pre, post, or during flights, and the HD video streaming gives a clearer and wider image than ever before. Furthermore, our most recent advancements have made it possible to alter the quality and latency optimisation mid-flight, just another customisable feature we provide our users with.

Sky-Drones Cloud

Cloud-based web application for drone fleet management, performance monitoring, mission planning, real-time control and post-flight AI analytics.

<!-- To be replaced with actual contents. Only for visual satisfaction now -->

Yoga

its origin Once known mainly as the “yoga of the stars,” the Bikram style of hatha has spread from Beverly Hills throughout the United States since the late 1970s. The Bikram style is the original “hot yoga” style, and its classes are taught in a room kept at approximately 106 degrees Fahrenheit (41 degrees Celsius). Bikram yoga is based on one series consisting of 26 poses,

Hot Yoga is a moving meditation combining asana, breath work and heat. All are needed to get the best result. My first yoga book I bought during my trip to Thailand has set off a series of events that truly made me who I am now. Believe it or not. 26 poses 1. PRANAYAMA (STANDING DEEP BREATHING) This is the beginning of each Bikram Yoga training. You breathe in warm air to heat up your body inside, you fill your blood with lots of oxygen, you loosen up our neck and shoulders and you are surprised about the funny sound of this breathing that fills the entire room. You try to find the rhythm with the teacher and the other students and before you can even think you have started the journey through your body in the here and now.

Drones

I remember the first time I held the drone rc at a flying school. The rotors whirled in a distance and echoed in my ears like thumping of my heart. I felt the exhilaration of the air vehicle taking off smoothly, balancing and firm in its position at my fingertips.

It goes back almost 40 years at my sophomore year when somehow I got a notion to apply for the airforce academy, which never got into a reality. Now I have this dream of flying UAVs all programmatically. This is a chapter of my life I started and would finish to the end until when satisfaction should be on my lips, in aftermath of all the turbulence in life.

Projects

Lorem ipsum dolor sit amet, consectetur adipisicing elit. Quae repudiandae fugiat illo cupiditate excepturi esse officiis consectetur, laudantium qui voluptatem. Ad necessitatibus velit, accusantium expedita debitis impedit rerum totam id. Lorem ipsum dolor sit amet, consectetur adipisicing elit. Natus quibusdam recusandae illum, nesciunt, architecto, saepe facere, voluptas eum incidunt dolores magni itaque autem neque velit in. At quia quaerat asperiores.

Learning

<!-- Service List -->
<!-- The circle icons use Font Awesome's stacked icon classes. For more information, visit http://fontawesome.io/examples/ -->

RAPA project

Final project with fusion filters and dronekit api based on Udacity platform github

Matlab Simulation

Simulink and other drone toolbox empowering APM simulation

ROS dronekit

ROS-bag and Gazebo

Delivery drone

Yolov3 Bounding Boxes: Here is where we created a transfer learning model from the Yolov3 architecture to find bounding boxes of cars, people, and trees in our images. These bounding boxes were used by our probability model to calculate the probability of collision.

Behavioral cloning

The model I trained can do infinitely number of laps at either 0.2, 0.25, or 0.3 throttle settings. However at 0.3 its execution on the first corner is not as smooth as I like (even though it doesn't go on to the concrete curve), so I decided to submit the drive.py with the 0.25 throttle setting.

Kalman filters

Medial Axis and Grid discretization have diagonal actions activated by default, you can change the value in the MotionPlanner constructor. Note that all the scripts use arguments to define the goal position. Use --goal_lon=x --goal_lat=y --goal_alt=z to use a custom destination. A default one is defined so it's not mandatory.

Jetbot gazebo

I have used RViz to create this plugin to be used in my Gazebo simulator.

Matlab Nvidia

Yuneec edison fc

Projects from previous work

freepbx

×Note: clone config repo and copy downloads/downloads and run the ‘install.sh’ to make a new server in sync with your nlp server

documents
cdc project check its submodule onoffjoin
freepbx install
develop journal check VSC journal workspace during this time
callconf based on wowza

root@159.65.8.44 check the nginx to reconfigure its gui dashboard

images

unity autonomous motor HUD

mobis collaboration with jotaekwan
project dashboard check 10 dashboard
mobis3 tony branch to sync with hardware PLC

With psd proposal design, I have worked on an album navigation control created in Unity

tony at work

이직준비
- 2019년 부터 근무해서 3년 경력이 생기고 2021년부터 2021.12까지 waiting!!
  방금 전(2022.1. 저녁)에도 합격 통보 하나 더 받은 상황
  - 가슴에 상처 많이 받았지만 이직 준비는 포기하지 말자
  - 기술스택이 다른 경우가 많았고, 규모의 차이는 어찌할 수 없었다고 생각하자
- 잘 되면 이직해서 좋고, 잘 못 되도 개인 역량 키우기엔 이만한게 없다.
  준비하며 배우는게 많았고, 조만간 따라잡을 수 있을 거 같다.
  - 기술과 언어들 어느정도 요령을 알았으니 올해는 해볼만 하다.
  - 개발자 포기하려고 수천번 생각했는데 포기안한게 용하다.
  - 2018.7 ~ 2020.05 3년반동안 거의 매일 퇴근길, 잠들기전
    하루평균 5~10번 포기하려고 생각했던것 같음
    - 1100 * 7.5 = 8250, 1만번 까지는 안한것 같다.
    - 포기하지 않게해달라고 하늘에 빈 것도 거의 그정도 되는 것 같다.
생각하는 범위까지는 개발이 가능하다.
- 생각의 범위를 넓혀야한다.
openapi로 블로그 포스팅하는 법을 익혔다.
- 제공해주는 openapi와 crawling 으로만으로 글을 쓰는데 현재는 하루 2000명 수준 방문(3개 블로그)
- 사람들이 관심사를 찾아내서 방문자를 더 늘려보자.
- 직접 글을 쓰는게 아니라 한번만 잘 만들어두면 하루에 1번씩 돌게 만들어서 자동 포스팅 할 수 있다.