RatSLAM: a hippocampal model for simultaneous localization and mapping

  title={RatSLAM: a hippocampal model for simultaneous localization and mapping},
  author={Michael Milford and Gordon Wyeth and David Prasser},
  journal={IEEE International Conference on Robotics and Automation, 2004. Proceedings. ICRA '04. 2004},
  pages={403-408 Vol.1}
The work presents a new approach to the problem of simultaneous localization and mapping - SLAM - inspired by computational models of the hippocampus of rodents. [] Key Method It uses a competitive attractor network to integrate odometric information with landmark sensing to form a consistent representation of the environment. Experimental results show that RatSLAM can operate with ambiguous landmark information and recover from both minor and major path integration errors.
Simultaneous localisation and mapping from natural landmarks using RatSLAM
The effectiveness of RatSLAM, a Simultaneous Localisation and Mapping system based on models of the rodent hippocampus, is shown in real robot tests in unmodified indoor environments using a learning vision system.
RatSLAM on the Edge: Revealing a Coherent Representation from an Overloaded Rat Brain
A new component for the RatSLAM system is described, which provides a coherent representation for goal directed navigation and the ability of the experience map to adapt to simple short term changes in the environment.
Effect of representation size and visual ambiguity on RatSLAM system performance
This paper describes experiments investigating the effects of the environment-representation size ratio and visual ambiguity on mapping and goal navigation performance and demonstrates that system performance is weakly dependent on either parameter in isolation, but strongly dependent on their joint values.
A Bio-Inspired Model of Navigation in a Multi-Chamber Maze
  • M. Bureš, M. Jirina
  • Biology, Psychology
    2006 IEEE International Conference on Computational Cybernetics
  • 2006
A model of spatial memory inspired by rodent hippocampus, which provides a robust and efficient method of storing several navigational maps in a single attractor neural (Hopfield-like) network, could provide a potential foundation for a future robotic use.
RatSLAM: Using Models of Rodent Hippocampus for Robot Navigation and Beyond
We describe recent biologically-inspired mapping research incorporating brain-based multi-sensor fusion and calibration processes and a new multi-scale, homogeneous mapping framework. We also review
Rapid Learning of Spatial Representations for Goal-Directed Navigation Based on a Novel Model of Hippocampal Place Fields
This paper develops a self-organized model incorporating place cells and replay, and demonstrates its utility for rapid one-shot learning in non-trivial environments with obstacles.
A Model of Navigation in a Complex Maze Inspired by Hippocampus
A model of navigation inspired by rodent hippocampus stores several navigational maps in a single attractor (Hopfield-like) neural network and structure of the model conforms to functional schema of hippocampal formation.
Biologically inspired visual landmark processing for simultaneous localization and mapping
A method for finding useful visual landmarks for performing simultaneous localization and mapping (SLAM) using layers of filtering and pooling to create learned templates that correspond to different views of the environment.
A Bionic Robot Navigation Algorithm Based on Cognitive Mechanism of Hippocampus
A bionic robot navigation algorithm based on the cognitive mechanism of rat hippocampus that makes mobile robots more intelligent by learning and imitating human and animal's environmental cognition and navigation ability is proposed.


Hippocampal models for simultaneous localisation and mapping on an autonomous robot
Tests in two dimensional environments revealed the inability of the RatSLAM system to maintain multiple pose hypotheses in the face of ambiguous visual input, support recent rat experimentation that suggest current competitive attractor models are not a complete solution to the hippocampal modelling problem.
Modeling rodent head-direction cells and place cells for spatial learning in bio-mimetic robotics
A computational model which is consistent with several neurophysiological findings concerning biological head-direction cells and hippocampal place cells is proposed, and the importance of correlating idiothetic and allothetic signals to determine the dynamics of the system in order to stabilize head- direction and place representations over time is stressed.
A modular neural architecture is put forward based on the functional properties as well as the anatomical interconnections of the brain regions involved in space representation that allows an artificial agent with animal-like exploration and self-localization capabilities to accomplish effective target-oriented navigation exploiting its interaction with the world.
A coupled attractor model of the rodent head direction system
A neural network model is described that creates a stable, distributed representation of head direction and updates that representation in response to angular velocity information and makes neurophysiological predictions that can be tested using current technologies.
Learning Maps for Indoor Mobile Robot Navigation.
By combining both paradigms-grid-based and topological-, the approach presented here gains the best of both worlds: accuracy/consistency and efficiency.
Probabilistic visual recognition of artificial landmarks for simultaneous localization and mapping
  • D. Prasser, G. Wyeth
  • Computer Science
    2003 IEEE International Conference on Robotics and Automation (Cat. No.03CH37422)
  • 2003
How uncertainty can be characterised for a vision system that locates coloured landmarks in a typical laboratory environment is described, and the implementation of the system on a laboratory robot is explained and experimental results show the coherence of the uncertainty model.
Self-organizing continuous attractor networks and path integration: two-dimensional models of place cells
It is shown for two related models how the representation of the two-dimensional space in the continuous attractor network of place cells could self-organize by modifying the synaptic connections between the neurons, and also how the place being represented can be updated by idiothetic signals in a neural implementation of path integration.
Anticipatory robot navigation by simultaneously localizing and building a cognitive map
  • Yoichiro Endo, R. Arkin
  • Computer Science
    Proceedings 2003 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2003) (Cat. No.03CH37453)
  • 2003
This paper presents a method for a mobile robot to construct and localize relative to a "cognitive map", where the cognitive map is assumed to be a representational structure that encodes both
Real time data association for FastSLAM
This paper presents a real-world implementation of FastSLAM, an algorithm that recursively estimates the full posterior distribution of both robot pose and landmark locations and presents an extension to Fast SLAM that addresses the data association problem using a nearest neighbor technique.
Nebof "Real Time Data Association fm Alf&+d
  • Conference on Robotics & Automation,
  • 1997