A Hybrid Controller based on the Egocentric Perceptual Principle

Rabinovich, Zinovi and Jennings, Nick (2010) A Hybrid Controller based on the Egocentric Perceptual Principle Robotics and Autonomous Systems, 58, (9), pp. 1038-1045.


[img] PDF ras.pdf - Accepted Manuscript
Download (197kB)
[img] PDF ROBOT1757.pdf - Version of Record
Download (1MB)


In this paper we extend the control methodology based on Extended Markov Tracking (EMT) by providing the control algorithm with capabilities to calibrate and even partially reconstruct the environment's model. This enables us to resolve the problem of performance deterioration due to model incoherence, a problem faced in all model-based control methods. The new algorithm, Ensemble Actions EMT (EA-EMT), utilises the initial environment model as a library of state transition functions and applies a variation of prediction with experts to assemble and calibrate a revised model. By so doing, this is the first hybrid control algorithm that enables on-line adaptation within the egocentric control framework which dictates the control of an agent's perceptions, rather than an agent's environment state. In our experiments, we performed a range of tests with increasing model incoherence induced by three types of exogenous environment perturbations: catastrophic -- the environment becomes completely inconsistent with the model, deviating -- some aspect of the environment behaviour diverges compared to that specified in the model, and periodic -- the environment alternates between several possible divergences. The results show that EA-EMT resolved model incoherence and significantly outperformed its EMT predecessor by up to 95%.

Item Type: Article
Keywords: hybrid control, perceptual control, dynamics based control, Kullback-Leibler divergence
Organisations: Agents, Interactions & Complexity
ePrint ID: 271139
Date :
Date Event
Date Deposited: 21 May 2010 11:42
Last Modified: 17 Apr 2017 18:21
Further Information:Google Scholar
URI: http://eprints.soton.ac.uk/id/eprint/271139

Actions (login required)

View Item View Item