同济大学 2023 年《计算机视觉》课程资料

.
├── CVBook
├── README.md
├── code
├── hw  # homework
├── pj  # project
├── ppt
└── ref

• Homework1

• Homework2

• Homework3

In this project, your task is to develop a simple interactive app between a mobile device and a ROS host, based on Android or IOS. The app should realize the following functions:

• 1. Start a SLAM application (Google’s cartographer 2D/3D) on the ROS master from the app. The program will play the data bag collected in advance, and then localize and map simultaneously.
• 2. The app needs to receive messages sent by the SLAM program and display them (3/6-DoF poses and the built map (2D grid map or 3D point cloud)) on its UI.
• 3. The app sends a command to early exit the SLAM program.

Caution: In the above-mentioned tasks, 2D is the basic requirement. Encouraging students to achieve 3D display can result in extra bonus.

• 1. Start a SLAM application (Google cartographer) in the ROS system.
• 2. Combined with hardware, self-purchased RPLIDAR A1M8 LiDAR to realize autonomous map building.
• 3. Use Sprint robosenes 16-wire LiDAR to generate a 3D point cloud map.
• 4. The cartographer accepts the radar data and sends it to the mobile app after processing.
• 5. Real-time synchronization between the app and cartographer.
• 6. Display of 2D grid map and 3D point cloud.
• 7. Switch between 2D and 3D maps on mobile.
• 8. Exit SLAM program from the app.

For details, please refer to pj/README.md

CVBook: https://github.com/csLinZhang/CVBook