The Labeled Multi-Bernoulli SLAM Filter

Deusch, Hendrik; Reuter, Stephan; Dietmayer, Klaus

doi:10.1109/lsp.2015.2414274

Cited by 86 publications

(52 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In some SLAM solutions, detection statistics are incorporated via the use of a binary Bayes filter to update each feature's probability of existence. Random Finite Set (RFS)‐based filters include probability of detection and false alarm statistics into the filter's update step, making the feature detector's detection statistics an intrinsic part of the Bayesian state estimation process. All of these techniques use estimates of the detection statistics, and this paper therefore provides principled methods for their estimation for use with RFS and vector‐based SLAM frameworks.…”

Section: Related Workmentioning

confidence: 99%

Modeling Detection Statistics in Feature‐Based Robotic Navigation for Range Sensors

Inostroza

Adams

Leung³

2018

J Inst Navig

View full text Add to dashboard Cite

This paper proposes using the number of range measurements that a detector utilizes to generate a detection as its descriptor. This one dimensional descriptor can be calculated with many range‐based detectors, and its expected value is used to derive detection statistics which take into account feature occlusions to improve robotic navigation performance. To demonstrate the advantages of estimating detection statistics, they are estimated and tested within Random Finite Set and vector‐based Simultaneous Localization and Mapping (SLAM) algorithms. Results from simulations and real experiments demonstrate the advantages of explicitly modeling feature detection statistics in both frameworks. © 2018 Institute of Navigation.

show abstract

Section: Related Workmentioning

confidence: 99%

Modeling Detection Statistics in Feature‐Based Robotic Navigation for Range Sensors

Inostroza

Adams

Leung³

2018

J Inst Navig

View full text Add to dashboard Cite

show abstract

“…40 Late breaking developments have also derived a solution for simultaneous localization and mapping (SLAM) based on propagating the distribution of the landmarks as an LMB density. 20 …”

Section: The Labeled Multi-bernoulli Filtermentioning

confidence: 99%

“…18 GLMB based filters have furthermore been adapted for partially observable Markov decision process (POMDP) based decision and control with closed form Cauchy-Schwarz divergence calculations 19 as well as for simultaneous localization and mapping (SLAM). 20 Indeed RFS based filtering solutions have also been applied in a wide range of areas from robust and approximate control, 21-27 visual tracking, 28-30 autonomous driving, 31-35 road safety, [36][37][38] to space situational awareness. [39][40][41][42] Further author information: E-mail: ba-ngu.vo@curtin.edu.au, ba-tuong.vo@curtin.edu,au…”

Section: Introductionmentioning

confidence: 99%

Random finite set multi-target trackers: stochastic geometry for space situational awareness

2015

SPIE Proceedings

View full text Add to dashboard Cite

This paper describes the recent development in the random finite set RFS paradigm in multi-target tracking. Over the last decade the Probability Hypothesis Density filter has become synonymous with the RFS approach. As result the PHD filter is often wrongly used as a performance benchmark for the RFS approach. Since there is a suite of RFS-based multi-target tracking algorithms, benchmarking tracking performance of the RFS approach by using the PHD filter, the cheapest of these, is misleading. Such benchmarking should be performed with more sophisticated RFS algorithms. In this paper we outline the high-performance RFS-based multi-target trackers such that the Generalized Labled Multi-Bernoulli filter, and a number of efficient approximations and discuss extensions and applications of these filters. Applications to space situational awareness are discussed.

show abstract

“…Classical probabilistic localization approaches have shown great results in unstructured environments enabling the successful deployment of autonomous vehicles in many cities around the world using either a grid-based representation of the environment [9] or a landmark-based map. In the latter case, the map contains static and easily recognizable objects [10], [11]. Storing objects in a landmarks-based map requires small amount of memory compared to the gridbased maps, where the area is discretized into small cells.…”

Section: Introductionmentioning

confidence: 99%

DeepLocalization: Landmark-based Self-Localization with Deep Neural Networks

Engel

Hoermann

Horn

et al. 2019

2019 IEEE Intelligent Transportation Systems Conference (ITSC)

Self Cite

View full text Add to dashboard Cite

We address the problem of vehicle selflocalization from multi-modal sensor information and a reference map. The map is generated off-line by extracting landmarks from the vehicle's field of view, while the measurements are collected similarly on the fly. Our goal is to determine the autonomous vehicle's pose from the landmark measurements and map landmarks. To learn this mapping, we propose DeepLocalization, a deep neural network that regresses the vehicle's translation and rotation parameters from unordered and dynamic input landmarks. The proposed network architecture is robust to changes of the dynamic environment and can cope with a small number of extracted landmarks. During the training process we rely on synthetically generated ground-truth. In our experiments, we evaluate two inference approaches in real-world scenarios. We show that DeepLocalization can be combined with regular GPS signals and filtering algorithms such as the extended Kalman filter. Our approach achieves state-of-the-art accuracy and is about ten times faster than the related work.

show abstract

The Labeled Multi-Bernoulli SLAM Filter

Cited by 86 publications

References 14 publications

Modeling Detection Statistics in Feature‐Based Robotic Navigation for Range Sensors

Modeling Detection Statistics in Feature‐Based Robotic Navigation for Range Sensors

Random finite set multi-target trackers: stochastic geometry for space situational awareness

DeepLocalization: Landmark-based Self-Localization with Deep Neural Networks

Contact Info

Product

Resources

About