Undergrad Thesis - An Exploration of Algorithms Enabling a Semantics-Aware Class-Based Probablistic Dynamic SLAM

This thesis describes the work that I performed persuant to recieving my undergraduate degree at CU Boulder. This thesis was advised by the exceptional Christoffer Heckman. In it, we explore utilizing a semantics aware clique-based persistence filter to handle non-static elements in a environment. Further, multi-agent tracking systems are reviewed and explored to provide semantic and panoptic segmentation of an environment to track persistance statistics of objects in a world online. These are used with the language of Survival Analysis to provide adaptive, high quality survival time priors to the persitence filter. Finally, usage of gaussian processes to estimate survival time prior distribution online is suggested using the devised framework, however, time restricted the exploration of the final goal of the paper.

The full thesis can be found here.