Gaussian Process Gauss–Newton for non-parametric simultaneous localization and mapping

Tong, Chi Hay; Furgale, Paul; Barfoot, Timothy D.

doi:10.1177/0278364913478672

Cited by 49 publications

(60 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recent work in the robotics [47] and computer vision [48] literature has shown promising results in this direction, but utilizes Gauss-Newton-based approaches that (as demonstrated herein) limit the class of objective functions that can be reliably employed. It would be interesting to consider applications of RISE (and related techniques) in the context of online machine learning in future work.…”

Section: Discussionmentioning

confidence: 99%

RISE: An Incremental Trust-Region Method for Robust Online Sparse Least-Squares Estimation

2014

View full text Add to dashboard Cite

Abstract-Many point estimation problems in robotics, computer vision and machine learning can be formulated as instances of the general problem of minimizing a sparse nonlinear sum-ofsquares objective function. For inference problems of this type, each input datum gives rise to a summand in the objective function, and therefore performing online inference corresponds to solving a sequence of sparse nonlinear least-squares minimization problems in which additional summands are added to the objective function over time. In this paper we present Robust Incremental least-Squares Estimation (RISE), an incrementalized version of the Powell's Dog-Leg numerical optimization method suitable for use in online sequential sparse least-squares minimization. As a trust-region method, RISE is naturally robust to objective function nonlinearity and numerical ill-conditioning, and is provably globally convergent for a broad class of inferential cost functions (twice-continuously differentiable functions with bounded sublevel sets). Consequently, RISE maintains the speed of current state-of-the-art online sparse least-squares methods while providing superior reliability.

show abstract

Section: Discussionmentioning

confidence: 99%

RISE: An Incremental Trust-Region Method for Robust Online Sparse Least-Squares Estimation

2014

View full text Add to dashboard Cite

show abstract

“…In an effort to improve representational power, Tong et al [3] use a Gaussian process to model the continuoustime trajectory of a robot. Although using a Gaussian process in essence provides an infinite resolution solution, we note that the often weak prior term, common to many of the continuous-time SLAM formulations including the one presented in this paper, plays a much more important role without a finite basis to help prevent overfitting.…”

Section: Related Workmentioning

confidence: 99%

“…Continuous-time batch estimation has been performed both parametrically, using a weighted sum of temporal basis functions [2], and non-parameterically, using a Gaussian-Process [3]. A benefit of the continuous-time approach is that it provides an elegant means to include asynchronous and high-rate measurements, such as those from an inertial measurement unit (IMU) or even individually timestamped features extracted from a motion-distorted image, into a typical batch nonlinear optimization framework.…”

Section: Introductionmentioning

confidence: 99%

A hierarchical wavelet decomposition for continuous-time SLAM

Anderson

Dellaert

Barfoot

2014

2014 IEEE International Conference on Robotics and Automation (ICRA)

Self Cite

View full text Add to dashboard Cite

Abstract-This paper proposes using hierarchical wavelets as a basis in parametric continuous-time batch estimation. The need for a continuous-time robot pose in the simultaneous localization and mapping (SLAM) problem has arisen as stateof-the-art batch SLAM algorithms attempt to handle more challenging hardware; specifically, the continuous-time framework is particularly beneficial when using high-rate sensors, multiple unsynchronized sensors, or scanning sensors, such as lidar and rolling-shutter cameras, during motion. Although the traditional discrete-time SLAM formulation can be adapted by using temporal pose interpolation, approaches using the continuous-time framework are able to generate smooth robot trajectories with less state variables. In this paper, we focus on the parametric approach using temporal basis functions to develop a finite-element representation of the continuous-time robot trajectory. While the majority of current implementations have utilized a uniformly spaced B-spline basis, we note that trajectory richness is often quite variable; in this paper, we show how a hierarchical system of wavelet basis functions can be used to increase the resolution of the solution only in the temporally local regions of the trajectory that require additional detail. We validate our approach by contrasting uniform Bsplines and wavelets in a six-dimensional pose-graph SLAM experiment, using both simulated and real data.

show abstract

“…Tong et al [23], based on Furgale et al framework [13], proposed a non-parametric representation for SLAM, where the state is represented as a Gaussian process. This is shown to perform better than discrete-time SLAM, albeit with increased computation time.…”

Section: Related Workmentioning

confidence: 99%

Continuous trajectory estimation for 3D SLAM from actuated lidar

Alismail

Baker

Browning

2014

2014 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

Abstract-We extend the Iterative Closest Point (ICP) algorithm to obtain a method for continuous-time trajectory estimation (CICP) suitable for SLAM from actuated lidar. Traditional solutions to SLAM from actuated lidar rely heavily on the accuracy of an auxiliary pose sensor to form rigid frames. These frames are then used with ICP to obtain accurate pose estimates. However, since lidar records a single range sample at time, any error in inter-sample sensor motion must be accounted for. This is not possible if the frame is treated as a rigid point cloud. In this work, instead of ICP we estimate a continuous-time trajectory that takes into account intersample pose errors. The trajectory is represented as a linear combination of basis functions and formulated as a solution to a (sparse) linear system without restrictive assumptions on sensor motion. We evaluate the algorithm on synthetic and real data and show improved accuracy in open-loop SLAM in comparison to state-of-the-art rigid registration methods.

show abstract

Gaussian Process Gauss–Newton for non-parametric simultaneous localization and mapping

Cited by 49 publications

References 29 publications

RISE: An Incremental Trust-Region Method for Robust Online Sparse Least-Squares Estimation

RISE: An Incremental Trust-Region Method for Robust Online Sparse Least-Squares Estimation

A hierarchical wavelet decomposition for continuous-time SLAM

Continuous trajectory estimation for 3D SLAM from actuated lidar

Contact Info

Product

Resources

About