gideon-98 / vslam-for-autonomous-driving Goto Github PK

The project’s goal is to develop a SLAM system which can successfully reconstruct trajectory on the KITTI visual odometry data set. A solution may be inspired and also based on published entries to the KITTI odometry benchmark. The report needs to very clearly indicate which parts are own innovations.

Outcomes:

▶ Development of a comprehensive SLAM method

▶ Use of grayscale or color image information

▶ Evaluation of algorithm performance using KITTI GT

▶ Optional: use of LIDAR and/or stereo information

Requirements: ▶The project should be documented in a report of maximum 50 pages

▶The report should clearly state who has contributed to each section.

▶An oral exam is conducted based on the report (pass/fail)).

▶The project should be demonstrated on May 12. Details follow later.✅

▶The report should be handed in on May 26 at 12:00.

Current Steps :

▶ Implement KITTI into exercise 4 DONE

▶ Incorporate exercise 2's solution into exercise 4

▶ Develop and implement loop closure

▶ More advanced VSLAM methods (ORB-SLAM etc.)

▶ Implement Real-time visualization of mapping (Nice-To-Have)