Active Information Acquisition with Mobile Robots
Nikolay A. Atanasov, Jerome Le Ny, Kostas Daniilidis, George Pappas

Citation
Nikolay A. Atanasov, Jerome Le Ny, Kostas Daniilidis, George Pappas. "Active Information Acquisition with Mobile Robots". Talk or presentation, October, 2015; Poster presented at the 2015 TerraSwarm Annual Meeting.

Abstract
We consider the problem of controlling mobile sensing systems in order to improve the accuracy and efficiency of information acquisition in scenarios such as environmental monitoring, search and rescue, surveillance and reconnaissance, and simultaneous localization and mapping (SLAM). A multi-sensor active information acquisition problem capturing the common characteristics of these scenarios is formulated. The goal is to design control policies for the sensors, which minimize the entropy of the estimation task conditioned on the future measurements. First, we provide a non-greedy centralized solution, which is computationally fast, since it exploits linearized sensing models, and memory efficient, since it exploits sparsity in the environment model. Next, we decentralize the control to achieve linear complexity in the number of sensors and provide suboptimality guarantees. The combination of these techniques is an effective and scalable approach for controlled information acquisition with multiple sensors. For example, when applied to the multi-robot active SLAM problem, our results enable a decentralized nonmyopic solution that exploits sparsity in the planning process. Finally, we present a formulation of the active information acquisition problem for discrete-state sensor and target models. This is particularly useful for active object recognition, classification, and hypothesis testing problems.

Electronic downloads

Internal. This publication has been marked by the author for TerraSwarm-only distribution, so electronic downloads are not available without logging in.

Citation formats

• HTML

  Nikolay A. Atanasov, Jerome Le Ny, Kostas Daniilidis, George
  Pappas. <a
  href="http://www.terraswarm.org/pubs/631.html"><i>Active
  Information Acquisition with Mobile
  Robots</i></a>, Talk or presentation,  October,
  2015; Poster presented at the <a
  href="http://terraswarm.org/conferences/15/annual"
  >2015 TerraSwarm Annual Meeting</a>.

• Plain text

  Nikolay A. Atanasov, Jerome Le Ny, Kostas Daniilidis, George
  Pappas. "Active Information Acquisition with Mobile
  Robots". Talk or presentation,  October, 2015; Poster
  presented at the <a
  href="http://terraswarm.org/conferences/15/annual"
  >2015 TerraSwarm Annual Meeting</a>.

• BibTeX

  @presentation{AtanasovLeNyDaniilidisPappas15_ActiveInformationAcquisitionWithMobileRobots,
      author = {Nikolay A. Atanasov and Jerome Le Ny and Kostas
                Daniilidis and George Pappas},
      title = {Active Information Acquisition with Mobile Robots},
      month = {October},
      year = {2015},
      note = {Poster presented at the <a
                href="http://terraswarm.org/conferences/15/annual"
                >2015 TerraSwarm Annual Meeting</a>.},
      abstract = {We consider the problem of controlling mobile
                sensing systems in order to improve the accuracy
                and efficiency of information acquisition in
                scenarios such as environmental monitoring, search
                and rescue, surveillance and reconnaissance, and
                simultaneous localization and mapping (SLAM). A
                multi-sensor active information acquisition
                problem capturing the common characteristics of
                these scenarios is formulated. The goal is to
                design control policies for the sensors, which
                minimize the entropy of the estimation task
                conditioned on the future measurements. First, we
                provide a non-greedy centralized solution, which
                is computationally fast, since it exploits
                linearized sensing models, and memory efficient,
                since it exploits sparsity in the environment
                model. Next, we decentralize the control to
                achieve linear complexity in the number of sensors
                and provide suboptimality guarantees. The
                combination of these techniques is an effective
                and scalable approach for controlled information
                acquisition with multiple sensors. For example,
                when applied to the multi-robot active SLAM
                problem, our results enable a decentralized
                nonmyopic solution that exploits sparsity in the
                planning process. Finally, we present a
                formulation of the active information acquisition
                problem for discrete-state sensor and target
                models. This is particularly useful for active
                object recognition, classification, and hypothesis
                testing problems.},
      URL = {http://terraswarm.org/pubs/631.html}
  }
  

Posted by Nikolay A. Atanasov on 30 Sep 2015.
Groups: services