Trajectory learning for human-robot scientific data collection

We propose an integrated learning and planning framework that leverages knowledge from a human user along with prior information about the environment to generate trajectories for scientific data collection. The proposed framework combines principles from probabilistic planning with uncertainty modeling through nonparametric Bayesian methods to refine trajectories for execution by autonomous vehicles. The resulting techniques allow for trajectories specified by a user to be modified for reduced risk of collision and increased reliability. We test our approach in the underwater ocean monitoring domain, and we show that the proposed framework reduces the risk of collision with ship traffic by as much as 51% for an autonomous underwater vehicle operating in ocean currents. This work provides insight into the tools necessary for combining human-robot interaction with autonomous navigation.

[1]  Gaurav S. Sukhatme,et al.  Towards marine bloom trajectory prediction for AUV mission planning , 2010, 2010 IEEE International Conference on Robotics and Automation.

[2]  J. Yamamoto,et al.  Properties and Applications of the Interpolation Variance Associated with Ordinary Kriging Estimates , 2008 .

[3]  David Silver,et al.  Learning to search: Functional gradient techniques for imitation learning , 2009, Auton. Robots.

[4]  Geoffrey A. Hollinger,et al.  Learning uncertainty models for reliable operation of Autonomous Underwater Vehicles , 2013, 2013 IEEE International Conference on Robotics and Automation.

[5]  Alexander F. Shchepetkin,et al.  The regional oceanic modeling system (ROMS): a split-explicit, free-surface, topography-following-coordinate oceanic model , 2005 .

[6]  Geoffrey A. Hollinger,et al.  Active planning for underwater inspection and the benefit of adaptivity , 2012, Int. J. Robotics Res..

[7]  Geoffrey A. Hollinger,et al.  Risk‐aware Path Planning for Autonomous Underwater Vehicles using Predictive Ocean Models , 2013, J. Field Robotics.

[8]  Alistair Reid,et al.  1-Point RANSAC for extended Kalman filtering: Application to real-time structure from motion and visual odometry , 2010 .

[9]  Neil D. Lawrence,et al.  Computationally Efficient Convolved Multiple Output Gaussian Processes , 2011, J. Mach. Learn. Res..

[10]  H. Zhang,et al.  OurOcean - An Integrated Solution to Ocean Monitoring and Forecasting , 2006, OCEANS 2006.

[11]  David R. Thompson,et al.  Spatiotemporal path planning in strong, dynamic, uncertain currents , 2010, 2010 IEEE International Conference on Robotics and Automation.

[12]  Kian Hsiang Low,et al.  Information-Theoretic Approach to Efficient Adaptive Path Planning for Mobile Robotic Environmental Sensing , 2009, ICAPS.

[13]  Gregory Dudek,et al.  Towards quantitative modeling of task confirmations in human-robot dialog , 2011, 2011 IEEE International Conference on Robotics and Automation.

[14]  Gaurav S. Sukhatme,et al.  USC CINAPS Builds bridges : observing and monitoring the southern california bight , 2010 .

[15]  Pierre F. J. Lermusiaux,et al.  Uncertainty estimation and prediction for interdisciplinary ocean dynamics , 2006, J. Comput. Phys..

[16]  John Shawe-Taylor,et al.  Introduction to the Special Topic on Grammar Induction, Representation of Language and Language Learning , 2011, J. Mach. Learn. Res..

[17]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.

[18]  W. Chadwick,et al.  Volcanic inflation measured in the caldera of Axial Seamount: Implications for magma supply and future eruptions , 2009 .

[19]  Gaurav S. Sukhatme,et al.  USC CINAPS Builds Bridges , 2010, IEEE Robotics & Automation Magazine.

[20]  John H. Reif,et al.  Complexity of the mover's problem and generalizations , 1979, 20th Annual Symposium on Foundations of Computer Science (sfcs 1979).

[21]  David Silver,et al.  Learning from Demonstration for Autonomous Navigation in Complex Unstructured Terrain , 2010, Int. J. Robotics Res..

[22]  Young-Ho Kim,et al.  Spatial Interpolation for Robotic Sampling: Uncertainty with Two Models of Variance , 2012, ISER.

[23]  J. Yamamoto.  An Alternative Measure of the Reliability of Ordinary Kriging Estimates , 2000 .

[24]  Jean-Claude Latombe,et al.  Robot motion planning , 1970, The Kluwer international series in engineering and computer science.

[25]  Robert E. Davis,et al.  Statistics for the evaluation and comparison of models , 1985 .

[26]  Emilio Frazzoli,et al.  Sampling-based algorithms for optimal motion planning , 2011, Int. J. Robotics Res..

[27]  Gaurav S. Sukhatme,et al.  Towards mixed-initiative, multi-robot field experiments: Design, deployment, and lessons learned , 2011, 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems.