HyperSLAM: A Generic and Modular Approach to Sensor Fusion and Simultaneous Localization And Mapping in Continuous-Time

Abstract

Within recent years, Continuous-Time Simultaneous Localization And Mapping (CTSLAM) formalisms have become subject to increased attention from the scientific community due to their vast potential in facilitating motioncorrected feature reprojection and direct unsynchronized multi-rate sensor fusion. They also hold the promise of yielding better estimates in traditional sensor setups (e.g. visual, inertial) when compared to conventional discretetime approaches. Related works mostly rely on cubic, C 2 -continuous, uniform cumulative B-Splines to exemplify and demonstrate the benefits inherent to continuous-time representations. However, as this type of splines gives rise to continuous trajectories by blending uniformly distributed SE 3 transformations in time, it is prone to under-or overparametrize underlying motions with varying volatility and prohibits dynamic trajectory refinement or sparsification by design. In light of this, we propose employing a more generalized and efficient non-uniform split interpolation method in R×SU 2 ×R 3 and commence with development of 'HyperSLAM', a generic and modular CTSLAM framework. The efficacy of our approach is exemplified in proof-ofconcept simulations based on a visual, monocular setup.

Ctrl-VIO: Continuous-Time Visual-Inertial Odometry for Rolling Shutter Cameras

“…Vandeportaele et al [19] add more control points where the angular velocity or linear acceleration is larger, and vice versa. Hug et al [15] recommend employing a more generic and non-uniform representation to prevent the under or over-parameterization. Admittedly, although the nonuniform formulation is critical for continuous-time trajectory estimation, it has only been employed in just a few batch offline applications, lacking implementation and verification in online odometry systems.…”

“…In terms of parameterizing 6-DoF continuous-time trajectories, most existing methods adopt a cubic, cumulative Bspline which offers local control property [11,[14][15][16][17]. In these approaches, control points of B-splines are often distributed uniformly.…”

“…However, in many applications, there is no prior knowledge about the complexity of the trajectory to be estimated. Consequently, due to the inability to dynamically adjust the distribution of control points, uniform B-splines are prone to under-or over-parameterization [15]. An appealing alternative is the use of non-uniform B-splines, which allows for dynamic control point distribution instead of a fixed frequency [15,19,20].…”