Robotic systems are nowadays capable of solving complex navigation tasks. However, their capabilities limited to the knowledge designer and consequently lack generalizability initially unconsidered situations. This makes deep reinforcement learning (DRL) especially interesting, as these algorithms promise a self-learning system only relying on feedback from environment. In this paper, we consid...

The underwater glider has the advantages of low power, long endurance and high accuracy. Micro-Electro-MechanicalSystem (MEMS) grade inertial sensors are more suitable for an underwater glider because of their low cost and small size. Models of MEMS sensor noises which include not only the white noises and random walk terms but also the bias instabilities of the sensor noises are analyzed. The ...

The Concordance-based Medial Axis Transform (CMAT) presented in this paper is a multiscale medial axis (MMA) algorithm that computes medial evidence from grey-level boundary measures. It differs from previous algorithms of this type in considering the symmetry of both boundary position and strength. The resultant MMA transform is not prone to the generation of spurious medial responses exhibite...

Navigation systems of the future will be integrated in a complex network of driver assistance systems, providing navigation data to several active and passive systems. With respect to this development, requirements concerning navigation accuracy and reliability grow rapidly; especially in urban areas with difficult navigation conditions navigation errors should be avoided carefully. Hence, syst...

In the light of rapidly growing repositories capturing the movement trajectories of people in spacetime, the need for trajectory compression becomes obvious. This paper argues for semantic trajectory compression (STC) as a means of substantially compressing the movement trajectories in an urban environment with acceptable information loss. STC exploits that human urban movement and its large–sc...

in this paper, we present the results of a hierarchical slam/gps/ins/wlfp sensor fusion to be used in navigation system devices. due to low quality of the inertial sensors, even a short-term gps failure can lower the integrated navigation performance significantly. in addition, in gps denied environments, most navigation systems need a separate assisting resource, in order to increase the avail...

The medial axis transform is valuable for shape representation as it is complete and captures part structure. However, its exact computation for arbitrary 3D models is not feasible. We introduce a novel algorithm to approximate the medial axis of a polyhedron with a dense set of medial points, with a guarantee that each medial point is within a specified tolerance from the medial axis. Given th...

• We present a spline-based medial axis transform representation of binary images. Our method achieves very high compression ratios while keeping quality high. has better performance than regular representation. show how our enables generating super-resolution allows shape manipulation and captures multiscale skeletons. Medial axes are well-known descriptors used for representing, manipulating,...