Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Continuous-Time Simultaneous-Localization and Mapping
The emerging paradigm of Continuous-Time Simultaneous Localization And Mapping (CTSLAM) has become a competitive alternative to conventional discrete-time approaches in recent times and holds the additional promise of fusing multimodal sensor setups in a truly generic manner, rendering its importance to robotic navigation and manipulation seminal. Based on our recent works, there are several possible and interesting extensions that are currently under consideration as student theses, spanning from research-oriented to engineering-oriented topics We are looking forward to individually discussing the available theses in greater detail in person.
Keywords: Asynchronous Sensor Fusion, Continuous-Time
Simultaneous Localization and Mapping
The challenging task of jointly mapping a robot's surroundings and estimating its ego-motion is referred to as SLAM in the literature. Well-performing SLAM algorithms, in turn, lie at the very foundation of modern robotics and are essential to other, higher-level applications such as autonomous navigation, object segmentation and obstacle avoidance. In the context of CTSLAM, motion estimates are parametrized as continuous functions, which, in contrast to conventional (discrete-time) SLAM approaches, allow state queries at arbitrary instances in time. Based in this property, continuous-time parametrizations not only expose vast potential in simplifying generic multi-rate sensor fusion and motion-corrected measurements but also accommodate unsynchronized sensor calibration and promise to even yield preciser estimates with respect to conventional discrete-time approaches and, thus, give rise to many research directions.
The challenging task of jointly mapping a robot's surroundings and estimating its ego-motion is referred to as SLAM in the literature. Well-performing SLAM algorithms, in turn, lie at the very foundation of modern robotics and are essential to other, higher-level applications such as autonomous navigation, object segmentation and obstacle avoidance. In the context of CTSLAM, motion estimates are parametrized as continuous functions, which, in contrast to conventional (discrete-time) SLAM approaches, allow state queries at arbitrary instances in time. Based in this property, continuous-time parametrizations not only expose vast potential in simplifying generic multi-rate sensor fusion and motion-corrected measurements but also accommodate unsynchronized sensor calibration and promise to even yield preciser estimates with respect to conventional discrete-time approaches and, thus, give rise to many research directions.