Limits...
Information-Driven Active Audio-Visual Source Localization.

Schult N, Reineking T, Kluss T, Zetzsche C - PLoS ONE (2015)

Bottom Line: These actions by the robot successively reduce uncertainty about the source's position.Because of the robot's mobility, this approach is suitable for use in complex and cluttered environments.We present qualitative and quantitative results of the system's performance and discuss possible areas of application.

View Article: PubMed Central - PubMed

Affiliation: Cognitive Neuroinformatics, Bremen University, Bremen, Germany.

ABSTRACT
We present a system for sensorimotor audio-visual source localization on a mobile robot. We utilize a particle filter for the combination of audio-visual information and for the temporal integration of consecutive measurements. Although the system only measures the current direction of the source, the position of the source can be estimated because the robot is able to move and can therefore obtain measurements from different directions. These actions by the robot successively reduce uncertainty about the source's position. An information gain mechanism is used for selecting the most informative actions in order to minimize the number of actions required to achieve accurate and precise position estimates in azimuth and distance. We show that this mechanism is an efficient solution to the action selection problem for source localization, and that it is able to produce precise position estimates despite simplified unisensory preprocessing. Because of the robot's mobility, this approach is suitable for use in complex and cluttered environments. We present qualitative and quantitative results of the system's performance and discuss possible areas of application.

No MeSH data available.


Entropy of the PDF estimate (simulation).When utilizing the IG procedure, the entropy of the PDF estimate decreases on average with each action, while the estimate of the source’s position gets more precise. While random movements also lead to a monotonic decrease of entropy, the comparison of the curves shows clearly that the IG procedure is reducing the remaining uncertainty more efficiently.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4556528&req=5

pone.0137057.g008: Entropy of the PDF estimate (simulation).When utilizing the IG procedure, the entropy of the PDF estimate decreases on average with each action, while the estimate of the source’s position gets more precise. While random movements also lead to a monotonic decrease of entropy, the comparison of the curves shows clearly that the IG procedure is reducing the remaining uncertainty more efficiently.

Mentions: The IG procedure is based on the minimization of the expected entropy of the PDF estimate. Thus, in order to evaluate the action-selection process, we calculated the entropy averaged over all experiments as a function of the number of actions performed by the system and compared the results to the system variant utilizing random movements. The results are presented in Fig 8. The entropy of the estimated PDF decreases with each action and the number of actions needed to achieve an precise estimate is minimized. By interpreting entropy as a measure of uncertainty, these results show that the system is actively trying to reduce uncertainty with respect to the position estimate. The average entropy is clearly reduced with each action when utilizing the IG procedure, while random movements rarely lead to drastic reductions of entropy and the impact of each action is smaller in general.


Information-Driven Active Audio-Visual Source Localization.

Schult N, Reineking T, Kluss T, Zetzsche C - PLoS ONE (2015)

Entropy of the PDF estimate (simulation).When utilizing the IG procedure, the entropy of the PDF estimate decreases on average with each action, while the estimate of the source’s position gets more precise. While random movements also lead to a monotonic decrease of entropy, the comparison of the curves shows clearly that the IG procedure is reducing the remaining uncertainty more efficiently.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4556528&req=5

pone.0137057.g008: Entropy of the PDF estimate (simulation).When utilizing the IG procedure, the entropy of the PDF estimate decreases on average with each action, while the estimate of the source’s position gets more precise. While random movements also lead to a monotonic decrease of entropy, the comparison of the curves shows clearly that the IG procedure is reducing the remaining uncertainty more efficiently.
Mentions: The IG procedure is based on the minimization of the expected entropy of the PDF estimate. Thus, in order to evaluate the action-selection process, we calculated the entropy averaged over all experiments as a function of the number of actions performed by the system and compared the results to the system variant utilizing random movements. The results are presented in Fig 8. The entropy of the estimated PDF decreases with each action and the number of actions needed to achieve an precise estimate is minimized. By interpreting entropy as a measure of uncertainty, these results show that the system is actively trying to reduce uncertainty with respect to the position estimate. The average entropy is clearly reduced with each action when utilizing the IG procedure, while random movements rarely lead to drastic reductions of entropy and the impact of each action is smaller in general.

Bottom Line: These actions by the robot successively reduce uncertainty about the source's position.Because of the robot's mobility, this approach is suitable for use in complex and cluttered environments.We present qualitative and quantitative results of the system's performance and discuss possible areas of application.

View Article: PubMed Central - PubMed

Affiliation: Cognitive Neuroinformatics, Bremen University, Bremen, Germany.

ABSTRACT
We present a system for sensorimotor audio-visual source localization on a mobile robot. We utilize a particle filter for the combination of audio-visual information and for the temporal integration of consecutive measurements. Although the system only measures the current direction of the source, the position of the source can be estimated because the robot is able to move and can therefore obtain measurements from different directions. These actions by the robot successively reduce uncertainty about the source's position. An information gain mechanism is used for selecting the most informative actions in order to minimize the number of actions required to achieve accurate and precise position estimates in azimuth and distance. We show that this mechanism is an efficient solution to the action selection problem for source localization, and that it is able to produce precise position estimates despite simplified unisensory preprocessing. Because of the robot's mobility, this approach is suitable for use in complex and cluttered environments. We present qualitative and quantitative results of the system's performance and discuss possible areas of application.

No MeSH data available.