🚘Sejong University Selly Project Developers in 2020🚘
교내 카페 '팬도로시'부터 AI센터까지 인도 보행으로 자율주행하는 배달로봇입니다.
This is an Autonomous-driving robot software which drives on pavement only. First test path is pavement in front of main gate, Sejong University along with road from 'cafe Pandorosi' to 'AI center'.
카메라 센서만을 사용한 vision 기반 자율주행 알고리즘 개발.
This is an auto-driving algorithm based on vision.
- autonomous-driving-vision : Image analysis to aviod obstacle:pushpin:
- road-segmentation : Segmentation for pavement driving
- object-detection : object detection to divide moving obstacle such as car and fixed object such as tree
- VSLAM : Visual SLAM for mapping and localization to let robot know where it is
- If you use Lidar sensor, you can use cartographer for SLAM
Junuary/2020 - June/2020
- Embedded : Jetson nano, Arduino
- Vision : Python3, Jupyter Notebook, Tensorflow 2.0
- Navigation : C++, ROS, Android studio
| Requirement | Description |
|---|---|
| ZED2 | camera sensor |
| ydlidar | Lidar sensor for test |
| ROS_melodic | We need melodic version of ROS, because of ubuntu |
| Jetson_Nano | Main embedded, which uses ubuntu 18.04 |
| python | v3.6 or higher |
| opencv | v4.1.1 or higher |
| tensorflow | v2.0 or higher |
| git | We follow github flow |
- Motor-control : Motor control with Motor driver
- Navigation : Suggest route from current position to purpose point with GPS and SLAM
- cartographer : 2D SLAM based on LIDAR sensor
- selly_vision : Proto Image analysis to aviod obstacle
- software : lidar, rasberry camera, arduino motor test codes and Practice code about auto-driving robot
- selly_motor : arduino ROS subscriber and jetson nano ROS publisher, which controll motor
navigation- selly_service
selly_motorControl