ROS-based competition car
For an educational robotics company, we need to develop a ROS-based competition robot to meet the needs of existing university/college courses.
In this project, the content is organized into five main sections: ROS basics and control, OpenCV, laser and depth cameras, deep learning and voice control, and model training for autonomous driving.
In this workspace, many packages implemented by ROS are listed in the screenshot.
ROS basics and control
command to control by keyboard.
rqt_graph for checking the correlation of topic and node.
OpenCV
it can implement many open source samples of opencv.
KCF tracker is a fast and robust object tracking algorithm that uses kernelized correlation filters and the Discrete Fourier Transform to efficiently follow objects in video frames.
Laser
The Robotic car can 'feel' the world now.
mapping and navigation are the basic of ROS.
depth cameras
link the camera in the ros through node.
depth camera on the way.
AR code is implemented in the Rviz environment.
pointcloud init
point cloud display
deep learning and voice control
YOLO3 implemented in the car by depth cam.
trained by own datasets through yolo pre-trained model.