Special Issue on Emerging Applications of Stochastic Geometry in Autonomous Robotics
A special issue of the IEEE Robotics and Automation Magazine.
The robust interpretation of an autonomous vehicle's environment, as well as its own position in that environment, underlies almost all autonomous robotic applications. This Simultaneous Localization And Map building (SLAM) problem requires a robust representation of the vehicle's surroundings (the map) in the presence of sensing/feature detection uncertainties such as false positives, missed detections and spatial errors. Significant research activity now exists in representing both measurements and the feature based SLAM map as a Random Finite Set (RFS), rather than the conventionally used random vector. This is not merely a triviality of representation. Recent research has shown that Finite Set Statistics (FISST), developed for data fusion and estimation with RFSs, when applied to sensor representations and SLAM, can eliminate the necessity of fragile map management and feature associationalgorithms. Robotic sensing, mapping and SLAM applications which use FISST in the form of the Probability Hypothesis Density (PHD), Cardinalized PHD (C-PHD) and Multi-Target, Multi-Bernoulli (MeMBer) filters have already demonstrated robust means of representing uncertain sensor data and maps, with the unique abilities of jointly tracking both object spatial and existence uncertainties.
The RFS representation of sensor data and robotic maps therefore provides a robust paradigm under which the true number of features, which have entered the field(s) of view of an autonomous vehicle's sensor(s), as well as their locations, can be jointly estimated in a Bayes optimal manner, while taking into account feature detection and false alarm probabilities.
Scope, Description, and More Information
This special issue calls for magazine style articles on the direct application of FISST to the issues of autonomous robotic sensing, mapping and navigation with diverse sensing techniques and in various environments. The robustness of the solutions presented should be demonstrated in the presence of sensing and sensor processing uncertainty. Comparisons with conventional vector based techniques are also encouraged. Topics of interest include, but are not limited to:
- Use of Sets for Sensor and Map Representations
- PHD Smoothing and Filtering Applications in Robotics
- Sensing, Mapping and SLAM in high clutter levels
- Multi-Vehicle SLAM
- Jointly incorporating object existence and spatial uncertainties into mapping and SLAM
- Metrics to determine full mapping and SLAM errors
- Detecting and Tracking Extended Targets
To submit a paper, go here.
|Call for Papers||September 24, 2012|
|Deadline for Paper Submission||March 10, 2013|
|First Review||April 10, 2013|
|Final Review||May 10, 2013|
Guest Editors: Martin Adams, Ba-Ngu Vo and Ronald Mahler
Santiago (Region Metropolitana Santiago), Chile
Department of Electrical and Computer Engineering
Perth, WA Austrailia
Ronald P. Mahler
Lockheed Martin Advanced Technology Laboratories
Senior Staff Research Scientist
Eagan, MN USA