Random Finite Sets for Robot Mapping and SLAM

Mullane, John; Vo, Ba-Ngu; Adams, Martin; Vo, Ba-Tuong

doi:10.1007/978-3-642-21390-8

Cited by 53 publications

(25 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We will provide a brief overview of the main steps in the RB-PHD filter [2,8]. Our main focus is on the importance weighting step, which we will cover in detail in Section III.…”

Section: B the Rb-phd Filtermentioning

confidence: 99%

“…The method in which this is performed is the focus of this paper, and we will explore this in greater details in section III. 5) Merging and Pruning of the Map: Gaussians with small weights are eliminated from the intensity function, while Gaussians that are close to each other are merged together [2,13]. This approximation is critical in limiting the computational requirement of the RB-PHD filter.…”

Section: B the Rb-phd Filtermentioning

confidence: 99%

“…Recently, a different representation has been introduced for feature-based maps using random finite sets (RFSs) [2], in which, random vectors for individual features (or landmarks) are placed in a set instead of being concatenated into a single vector (as in the vector-based approach). The cardinality of the set is also a random variable.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

An improved weighting strategy for Rao-Blackwellized Probability Hypothesis Density simultaneous localization and mapping

Leung

Inostroza

Adams

2013

2013 International Conference on Control, Automation and Information Sciences (ICCAIS)

Self Cite

View full text Add to dashboard Cite

The use of random finite sets (RFSs) in simultaneous localization and mapping (SLAM) for mobile robots is a new concept that provides several advantages over traditional vector-based approaches. These include: 1) the incorporation of detection statistics, as well as the usual spatial uncertainty, in an estimation algorithm, 2) the ability to estimate the number of landmarks in a map, and 3) the circumvention of the need for data association heuristics. Solutions to SLAM can be obtained through the Rao-Blackwellized Probability Hypothesis Density (RB-PHD) filter, which is an approximation of the Bayes filter for RFSs using both particles to represent the robot trajectories, and Gaussian mixtures to represent their associated maps. This paper proposes an improved multi-feature particle weighting strategy for the RB-PHD filter and shows through simulations that it outperforms existing weighting strategies. The proposed strategy makes the RB-PHD filter a generalization of multi-hypothesis (MH) FastSLAM, a vector-based SLAM solution that uses the RB-particle filter.

show abstract

“…We will provide a brief overview of the main steps in the RB-PHD filter [2,8]. Our main focus is on the importance weighting step, which we will cover in detail in Section III.…”

Section: B the Rb-phd Filtermentioning

confidence: 99%

Section: B the Rb-phd Filtermentioning

confidence: 99%

See 1 more Smart Citation

An improved weighting strategy for Rao-Blackwellized Probability Hypothesis Density simultaneous localization and mapping

Leung

Inostroza

Adams

2013

2013 International Conference on Control, Automation and Information Sciences (ICCAIS)

Self Cite

View full text Add to dashboard Cite

show abstract

“…This treatment is necessitated by the concept of estimation error for multiple targets [16]. Over the last decade, RFS-based solutions such as the Probability Hypothesis Density (PHD) [9]- [11], Cardinalized Y. Punchihewa and F. Papi [12], [13], and multi-Bernoulli [14]- [16] has been successfully applied to many areas including radar/sonar [17]- [21], robotics [22]- [26], automotive safety [27]- [29], and biomedical research [30]- [32].…”

Section: Introductionmentioning

confidence: 99%

Multiple target tracking in video data using labeled random finite set

Punchihewa

Papi

Hoseinnezhad

2014

The 2014 International Conference on Control, Automation and Information Sciences (ICCAIS 2014)

View full text Add to dashboard Cite

This paper demonstrates how the δ-Generalized Labeled Multi-Bernoulli (δ-GLMB) filter can be applied to track moving targets on videos. The tracking is performed directly on the original images which are not preprocessed into point measurements and estimates the number of targets on frame along with their states. In that sense this concept bears resemblance to the track before detect (TBD) approach employed under low signal to noise ratio conditions. Image sequences from the CAVIAR 1 dataset are used in simulations to prove the aptitude of this method.

show abstract

“…An RFS is a random variable with realizations that are finite sets, jointly modeling the number of the targets as the size of the set and the state of each target as the individual elements of the set. Estimation algorithms based on RFSs are becoming increasingly popular and have been used in applications such as vehicle tracking [6], Simultaneous Localization and Mapping (SLAM) [7], robot localization [8], and target search [9]- [11].…”

Section: Introductionmentioning

confidence: 99%

Active Magnetic Anomaly Detection Using Multiple Micro Aerial Vehicles

Dames

Schwager

Rus

et al. 2016

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

Abstract-Magnetic Anomaly Detection (MAD) is an important problem in applications ranging from geological surveillance to military reconnaissance. MAD sensors detect local disturbances in the magnetic field, which can be used to detect the existence of and to estimate the position of buried, hidden, or submerged objects, such as ore deposits or mines. These sensors may experience false positive and false negative detections and, without prior knowledge of the targets, can only determine proximity to a target. The uncertainty in the sensors, coupled with a lack of knowledge of even the existence of targets, makes the estimation and control problems challenging. We utilize a hierarchical decomposition of the environment, coupled with an estimation algorithm based on random finite sets, to determine the number of and the locations of targets in the environment. The small team of robots follow the gradient of mutual information between the estimated set of targets and the future measurements, locally maximizing the rate of information gain. We present experimental results of a team of quadrotor micro aerial vehicles discovering and localizing an unknown number of permanent magnets.

show abstract

Random Finite Sets for Robot Mapping and SLAM

Abstract: The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

Cited by 53 publications

References 0 publications

An improved weighting strategy for Rao-Blackwellized Probability Hypothesis Density simultaneous localization and mapping

An improved weighting strategy for Rao-Blackwellized Probability Hypothesis Density simultaneous localization and mapping

Multiple target tracking in video data using labeled random finite set

Active Magnetic Anomaly Detection Using Multiple Micro Aerial Vehicles

Contact Info

Product

Resources

About