Learning Place Cells from Sonar Data
نویسندگان
چکیده
A place cell system is developed that is able to provide robust localisation using simple sonar information only. The system uses egocentric view cells, based on the Adaptive Response Function Neuron, as sensory input to correct for path integration errors. The major advantage of this system is that place field centres are fixed prior to training, allowing downstream navigational systems to make relevant predictions immediately upon entering a new environment.
منابع مشابه
A Framework for Autonomous Mobile Robot Exploration and Map Learning through the use of Place-Centric Occupancy Grids
The Autonomous Intelligent Knowledgebuilding Exploration (AIKE) system explores how an autonomous mobile robotic agent using sonar sensors and dead reckoning can learn a map of a completely unknown environment. AIKE generates grid-based maps based on the notion of a "place" (i.e., a location separated from other places by a confining gateway such as a door or entryway). Empirical trials show th...
متن کاملComparison of Learned versus Engineered Features for Classification of Mine Like Objects from Raw Sonar Images
Advances in high frequency sonar have provided increasing resolution of sea bottom objects, providing higher fidelity sonar data for automated target recognition tools. Here we investigate if advanced techniques in the field of visual object recognition and machine learning can be applied to classify mine-like objects from such sonar data. In particular, we investigate if the recently popular S...
متن کاملLearning to Avoid Objects and Dock·with a Mobile Robot
In this paper we describe a novel robot learning method that enables a mobile robot equipped with sonar and IR light sensors to automatically acquire the ability to negotiate objects and dock by simply interacting with the environment. We achieve this by providing the robot with sonar and IR sensors for detecting objects and the relative direction of IR beacons placed in the environment. A set ...
متن کاملSupervised Learning of Probability Distributions by Neural Networks
We propose that the back propagation algorithm for supervised learning can be generalized, put on a satisfactory conceptual footing, and very likely made more efficient by defining the values of the output and input neurons as probabilities and varying the synaptic weights in the gradient direction of the log likelihood, rather than the 'error'. In the past thirty years many researchers have st...
متن کاملGlobal localization and topological map-learning for robot navigation
This paper describes a navigation system implemented on a real mobile robot. Using simple sonar and visual sensors, it makes possible the autonomous construction of a dense topological map representing the environment. At any time during the mapping process, this system is able to globally localize the robot, i.e. to estimate the robot’s position even if the robot is passively moved from one pl...
متن کامل