Skip to content
This repository has been archived by the owner on Oct 26, 2020. It is now read-only.

Latest commit

 

History

History
46 lines (36 loc) · 2.88 KB

README.md

File metadata and controls

46 lines (36 loc) · 2.88 KB

Alpha Pilot: Team DeadLock

Introduction

This repo contains all the code that Team Deadlock developed in the alpha pilot competition. Developing a vision-based autonomous flying system is challenging, and we are proud of building this prototype from scratch in 3 months. Now, all the trails and failures we had are open-sourced. This hopefully can provide some hints for future participants.

Tasks for qualifier

We break down the large system into multiple smaller sub-systems by studying autonomous driving systems. In general, we need vision system to detect gates and estimate depth, control system to send commands to quadcopter, simulation to verify our system, SLAM system to localize our quadcopter in noisy sensor measurement.

Finally, we assmeble different systems in the e2e_system folder. Most of the code are in Python, considering we need to build a prototype fast.

Detail system breakdown:

State machine design

state machine diagram

Final reports

Team collaboration tool

We use Trello for weekly work planning, slack for daily communication and hangout for meetings.

Contribution guide

If you are working on a subcomponent, feel free to commit to the corresponding folder directly. If you are trying to modify the integration folder, please code in another branch and submit a pull request. Something to pay attention before submitting your code.

  • Remove unncessary data (especially large files) to avoid exceeding repo limit (1 GB)
  • When you are using a third party package, either specify its version (e.g. tensorflow==1.12) or bring in the source code (C++ project, unless it is under GPL license).