Aprendizagem de Máquina

Biologically-inspired robot localization (Place cells)

Submitted by erantone
on Wed, 12/09/2015 - 17:21

This work proposes a hierarchical biologically-inspired architecture for learning sensor-based spatial representations of a robot environment in an unsupervised way. The first layer is comprised of a fixed randomly generated recurrent neural network, the reservoir, which projects the input into a high-dimensional, dynamic space. The second layer learns instantaneous slowly-varying signals from the reservoir states using Slow Feature Analysis (SFA), whereas the third layer learns a sparse coding on the SFA layer using Independent Component Analysis (ICA). While the SFA layer generates non-localized activations in space, the ICA layer presents high place selectivity, forming a localized spatial activation, characteristic of place cells found in the hippocampus area of the rodent’s brain. We show that, using a limited number of noisy short-range distance sensors as input, the proposed system learns a spatial representation of the environment which can be used to predict the actual location of simulated and real robots, without the use of odometry. The results confirm that the reservoir layer is essential for learning spatial representations from low-dimensional input such as distance sensors. The main reason is that the reservoir state reflects the recent history of the input stream. Thus, this fading memory is essential for detecting locations, mainly when locations are ambiguous and characterized by similar sensor readings.

Video for data generation:

 

 

Publications

  1. Eric Antonelo and Benjamin Schrauwen Learning slow features with reservoir computing for biologically-inspired robot localization NEURAL NETWORKS, pp. 178-190 (2011)   
  2. Eric Antonelo and Benjamin Schrauwen Towards autonomous self-localization of small mobile robots using reservoir computing and slow feature analysis IEEE International conference on Systems, Man, and Cybernetics, Conference digest, Vol. 2, pp. (2009)   
  3. Eric Antonelo and Benjamin Schrauwen Unsupervised learning in reservoir computing : modeling hippocampal place cells for small mobile robots LECTURE NOTES IN COMPUTER SCIENCE, Vol. 5768, pp. 747-756 (2009)   

 

Cognitive computation for Deviation detection in Fleet of City Buses

Submitted by erantone
on Wed, 12/09/2015 - 16:21

With Prof. Thorsteinn Rögnvaldsson, from Halmstad University, Sweden, we are looking at how Reservoir Computing can help in deviation detection in a fleet of Swedish city buses using a signal from the air tank pressure from the buses in order to predict when a bus is going to break well in advance.

Video from the project at Halmstad University:

 

 

 

Related Publications: 

Predicting Air Compressor Failures with Echo State Networks

Submitted by erantone
on Fri, 09/30/2016 - 20:05
Fan Y, Nowaczyk S, Rögnvaldsson T, Antonelo EA.  2016.  Predicting Air Compressor Failures with Echo State Networks. Third European Conference of the Prognostics and Health Management Society 2016, Bilbao, Spain, 5-8 July, 2016.

Proxy dynamical models of offshore oil production platforms via recurrent neural networks

Submitted by erantone
on Wed, 12/09/2015 - 14:52

Process measurements are of vital importance for monitoring and control of industrial plants. When we consider offshore oil production platforms, wells that require gas-lift technology to yield oil production from low pressure oil reservoirs can become unstable under some conditions. This undesirable phenomenon is usually called slugging flow, and can be identified by an oscillatory behavior of the downhole pressure measurement.

Given the importance of this measurement and the unreliability of the related sensor, this work aims at designing data-driven soft-sensors for downhole pressure estimation in two contexts: one for speeding up first-principled model simulation of a vertical riser model; and another for estimating the downhole pressure using real-world data from an oil well from Petrobras based only on topside platform measurements. Both tasks are tackled by employing Echo State Networks (ESNs) as an efficient technique for training Recurrent Neural Networks.

We show that a single ESN is capable of robustly modeling both the slugging flow behavior and a steady state based only on a square wave input signal representing the production choke opening in the vertical riser. Besides, we compare the performance of a standard network to the performance of a multiple timescale hierarchical architecture in the second task and show that for some periods the latter architecture performs better. 

 

                    

 

Related Publications: 

An Echo State Network-based Soft Sensor of Downhole Pressure for a Gas-lift Oil Well

Submitted by erantone
on Thu, 09/29/2016 - 19:51
Antonelo EA, Camponogara E.  2015.  An Echo State Network-based Soft Sensor of Downhole Pressure for a Gas-lift Oil Well. 16th International Conference on Engineering Applications of Neural Networks. Communications in Computer and Information Science, vol 517:379-389.

Delayed Response Tasks in Robot Control

Submitted by erantone
on Wed, 12/09/2015 - 14:38

In this work we tackle the road sign problem with Reservoir Computing (RC) networks. The T-maze task (a particular form of the road sign problem) consists of a robot in a T-shaped environment that must reach the correct goal (left or right arm of the T-maze) depending on a previously received input sign. It is a control task in which the delay period between the sign received and the required response (e.g., turn right or left) is a crucial factor. Delayed response tasks like this one form a temporal problem that can be handled very well by RC networks. Reservoir Computing is a biologically plausible technique which overcomes the problems of previous algorithms such as Backpropagation Through Time - which exhibits slow (or non-) convergence on training. RC is a new concept that includes a fast and efficient training algorithm. We show that this simple approach can solve the T-maze task efficiently.

 

Video showing trained RC network controlling the robot:

 

Publications

  1. Eric Antonelo, Benjamin Schrauwen and Dirk Stroobandt Mobile Robot Control in the Road Sign Problem using Reservoir Computing Networks Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), pp. 911-916 (2008)    
  2. Eric Antonelo, Benjamin Schrauwen and Jan Van Campenhout Generative Modeling of Autonomous Robots and their Environments using Reservoir Computing Neural Processing Letters, Vol. 26(3), pp. 233-249 (2007)   

Event detection and localization for small mobile robots using reservoir computing

Submitted by erantone
on Wed, 12/09/2015 - 14:06

Reservoir Computing (RC) techniques use a fixed (usually randomly created) recurrent neural network, or more generally any dynamic system, which operates at the edge of stability, where only a linear static readout output layer is trained by standard linear regression methods. In this work, RC is used for detecting complex events in autonomous robot navigation. This can be extended to robot localization tasks which are solely based on a few low-range, high-noise sensory data. The robot thus builds an implicit map of the environment (after learning) that is used for efficient localization by simply processing the stream of distance sensors. These techniques are demonstrated in both a simple simulation environment and in the physically realistic Webots simulation of the commercially available e-puck robot, using several complex and even dynamic environments.

 

Videos showing data generation for event detection and localization:

 

 

Publications

  1. Eric Antonelo, Benjamin Schrauwen and Dirk Stroobandt Event detection and localization for small mobile robots using reservoir computing NEURAL NETWORKS, Vol. 21(6), pp. 862-871 (2008)  
  2. Eric Antonelo, Benjamin Schrauwen, Xavier Dutoit, Dirk Stroobandt and Marnix Nuttin Event detection and localization in mobile robot navigation using reservoir computing Proceedings of the International Conference on Artificial Neural Networks (ICANN), pp. 660-669 (2007)    

Pages

Powered by Drupal