Simultaneous Localisation And Mapping (SLAM) methods have been developed with the aim of automating robot navigation. Place recognition (or loop closure detection) is considered a key element, complementary to sequential motion estimation done in visual odometry/SLAM, to enable global accurate maps, relocalization and even collaboration between different robots performing SLAM (see image above). Place recognition is a challenging task, due to the large variability in a scene's appearance that can be observed in the real world, caused by changes in illumination, seasons, or presence of occlusions and dynamic objects. Conversely, different locations can appear identical. This is called perceptual aliasing and it is a key problem in Vision for Robotics attracting great research interest. A common approach in visual place recognition is to create “locations“ based on local similarity of neighbouring images and store these locations in a database, indexed by the image content/features that appears in them. When this database is queried at runtime to identify which location matches, the current best location with the most similar image content is returned. The goal of this project is to develop a method to return not only the most similar location, but also an estimation of the error in the position of the current image with respect to the loop-closure location. The novel method will be integrated in a state of the art visual-inertial odometry system to enable robust relocalization and more accurate maps. The student taking this project needs to be highly motivated, preferably with strong analytical skills, while experience in coding in C/C++ would be beneficial. The student will have the opportunity to work with a real setup and equipment offered by the Vision for Robotics Lab. This work is part of a large European project and a successful method will directly be used within in the framework developed for this project.
- WP1: Research into existing works tackling the problem of place recognition in SLAM.
- WP2: Development of a new algorithm to estimate the error in the position of the detected loop closure with relation to the current location.
- WP3: Experimentation and evaluation of this method in terms of runtime and accuracy of estimation.
- WP4: Further optimization of the pipeline to correct the map generated by the system (problem known as loop closing).
- WP5: Experimentation and final evaluation of the method with respect to the state of the art and report writing.
- Background knowledge in computer vision.
- C++ programming experience.
- Experience with Linux, ROS are advantageous.
- Strong self-motivation and critical mind.
Interested student please contact Fabiola Maffra (email@example.com).
CLS Student Project (MPG ETH CLS)
Information, Computing and Communication Sciences
Engineering and Technology