Search papers, labs, and topics across Lattice.
The paper introduces PathSpace, a semantic SLAM framework that uses continuous B-splines to represent environments compactly while maintaining probability density functions for probabilistic reasoning. This approach allows for the interpolation and fitting of sparse environments, addressing the limitations of dense geometric representations in current semantic SLAM techniques. Experiments in autonomous racing demonstrate that PathSpace achieves comparable accuracy to landmark-based methods with significantly reduced resource usage by exploiting pre-specified track characteristics.
Ditch the dense point clouds: PathSpace slashes SLAM's computational cost by representing environments as smooth B-splines, enabling efficient reasoning in constrained spaces.
Simultaneous Localization and Mapping (SLAM) plays a crucial role in enabling autonomous vehicles to navigate previously unknown environments. Semantic SLAM mostly extends visual SLAM, leveraging the higher density information available to reason about the environment in a more human-like manner. This allows for better decision making by exploiting prior structural knowledge of the environment, usually in the form of labels. Current semantic SLAM techniques still mostly rely on a dense geometric representation of the environment, limiting their ability to apply constraints based on context. We propose PathSpace, a novel semantic SLAM framework that uses continuous B-splines to represent the environment in a compact manner, while also maintaining and reasoning through the continuous probability density functions required for probabilistic reasoning. This system applies the multiple strengths of B-splines in the context of SLAM to interpolate and fit otherwise discrete sparse environments. We test this framework in the context of autonomous racing, where we exploit pre-specified track characteristics to produce significantly reduced representations at comparable levels of accuracy to traditional landmark based methods and demonstrate its potential in limiting the resources used by a system with minimal accuracy loss.
