DA-BSP: Towards Data Association Aware Belief Space Planning for Robust Active Perception

We develop a belief space planning (BSP) approach that advances the state of the art by incorporating reasoning about data association (DA) within planning (existing BSP approaches typically assume data association is given and perfect), while considering additional sources of uncertainty. Our data association aware belief space planning (DA-BSP) approach explicitly reasons about DA within belief evolution, and as such can better accommodate these challenging real world scenarios. Starting from a Gaussian prior, due to perceptual aliasing, we show that the posterior belief becomes a Gaussian mixture model. Overall, our approach is applicable to robust active perception and autonomous navigation in perceptually aliased environments.

[1]  Ryan M. Eustice,et al.  Risk aversion in belief-space planning under measurement acquisition uncertainty , 2015, 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[2]  Ryan M. Eustice,et al.  Belief space planning for underwater cooperative localization , 2015, 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[3]  Frank Dellaert,et al.  Incremental smoothing and mapping , 2008 .

[4]  Nathan Michael,et al.  Multi-robot pose graph localization and data association from unknown initial relative poses via expectation maximization , 2014, 2014 IEEE International Conference on Robotics and Automation (ICRA).

[5]  Leslie Pack Kaelbling,et al.  Planning and Acting in Partially Observable Stochastic Domains , 1998, Artif. Intell..

[6]  Ryan M. Eustice,et al.  Active visual SLAM for robotic area coverage: Theory and experiment , 2015, Int. J. Robotics Res..

[7]  George J. Pappas,et al.  Active Object Recognition via Monte Carlo Tree Search , 2015 .

[8]  Ryan M. Eustice,et al.  Opportunistic sampling-based planning for active visual SLAM , 2014, 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[9]  N. Roy,et al.  The Belief Roadmap: Efficient Planning in Belief Space by Factoring the Covariance , 2009, Int. J. Robotics Res..

[10]  Saurav Agarwal,et al.  Motion Planning in Non-Gaussian Belief Spaces for Mobile Robots , 2015, ArXiv.

[11]  Frank Dellaert,et al.  Planning in the continuous domain: A generalized belief space approach for autonomous navigation in unknown environments , 2015, Int. J. Robotics Res..

[12]  George J. Pappas,et al.  Nonmyopic View Planning for Active Object Classification and Pose Estimation , 2014, IEEE Transactions on Robotics.

[13]  Shashank Pathak,et al.  Robust Active Perception via Data-association aware Belief Space planning , 2016, ArXiv.

[14]  Ron Alterovitz,et al.  Motion planning under uncertainty using iterative local optimization in belief space , 2012, Int. J. Robotics Res..

[15]  Nathan Michael,et al.  Incremental Distributed Inference from Arbitrary Poses and Unknown Data Association: Using Collaborating Robots to Establish a Common Reference , 2016, IEEE Control Systems.

[16]  Leslie Pack Kaelbling,et al.  Data association for semantic world modeling from partial views , 2015, Int. J. Robotics Res..

[17]  Niko Sünderhauf,et al.  Towards a robust back-end for pose graph SLAM , 2012, 2012 IEEE International Conference on Robotics and Automation.

[18]  Edwin Olson,et al.  Inference on networks of mixtures for robust robot mapping , 2012, Robotics: Science and Systems.

[19]  Stefan Schaal,et al.  Policy Learning with Hypothesis based Local Action Selection , 2015, ArXiv.

[20]  Frank Dellaert,et al.  Selecting good measurements via ℓ1 relaxation: A convex approach for robust estimation over graphs , 2014, 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems.