A GA based SLAM with range sensors only

Wang, Han; Yan, Yi; Wang, Dan Wei

doi:10.1109/icarcv.2010.5707965

Cited by 3 publications

(2 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are different approaches in literature to achieve SLAM. Some of them are based on (extended) Kalman filters ( EKF) [ 4] [ 5], p article fi lters [6 ] or ge netic algorithms [7], and various variations on these. Genetic algorithm based SLAM [7] [8] [9] approaches are versatile, and they require very little prior information of the system, making them an acceptable choice for the system to be developed, thus in this paper genetic algorithm based SLAM is selected for localization and mapping.…”

Section: Introductionmentioning

confidence: 99%

Localization and 3D Mapping using 1D Automotive Radar Sensor

Gaalen

Uysal

Yarovoy

2020

2020 IEEE Radar Conference (RadarConf20)

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

Localization and 3D Mapping using 1D Automotive Radar Sensor

Gaalen

Uysal

Yarovoy

2020

2020 IEEE Radar Conference (RadarConf20)

View full text Add to dashboard Cite

“…Some biological evolution algorithms have been newly proposed to keep particle diversity as long as possible. 17,18 On the other hand, FastSLAM linearizes the motion model in the same manner as EKF-SLAM. Inaccurate approximation of the nonlinear function leads to filter divergence.…”

Section: Introductionmentioning

confidence: 99%

An improved FastSLAM 2.0 algorithm based on FC&ASD-PSO

Feng²

2016

Robotica

View full text Add to dashboard Cite

SUMMARYFastSLAM 2.0 is a popular framework which uses a Rao-Blackwellized particle filter to solve the simultaneous localization and mapping problem. The sampling process is one of the most important phases in the FastSLAM 2.0 framework. Its estimation accuracy depends heavily on a correct prior knowledge about the control and observation noise statistics (the covariance matrices Q and R). Without the correct prior knowledge about these matrices, the estimation accuracy of the robot path and landmark positions may degrade seriously. However in many applications, the prior knowledge is unknown, or these noises are non-stationary. In this paper, these covariance matrices are supposed to be dynamic and denoted as Qt and Rt. Since there are noises, time-adjacent observations are inconsistent with each other. This inconsistency can reflect the real value of the covariance matrices. By the inconsistency, an extra step is introduced to the FastSLAM 2.0 framework. This step makes Qt and Rt match with their real value by using a particle swarm optimization method based on fractional calculus and alpha stable distribution (FC&ASD-PSO). Both simulation and experimental results show that the proposed algorithm improves the accuracy by the more accurate estimation on the noise covariance matrices.

show abstract

A Fast Genetic SLAM Approach for Mobile Robots

Chen

Lee

et al. 2013

2013 14th ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed C

View full text Add to dashboard Cite

A GA based SLAM with range sensors only

Cited by 3 publications

References 11 publications

Localization and 3D Mapping using 1D Automotive Radar Sensor

Localization and 3D Mapping using 1D Automotive Radar Sensor

An improved FastSLAM 2.0 algorithm based on FC&ASD-PSO

A Fast Genetic SLAM Approach for Mobile Robots

Contact Info

Product

Resources

About