Risk aversion in belief-space planning under measurement acquisition uncertainty

This paper reports on a Gaussian belief-space planning formulation for mobile robots that includes random measurement acquisition variables that model whether or not each measurement is actually acquired. We show that maintaining the stochasticity of these variables in the planning formulation leads to a random belief covariance matrix, allowing us to consider the risk associated with the acquisition in the objective function. Inspired by modern portfolio theory and utility optimization, we design objective functions that are risk-averse, and show that risk-averse planning leads to decisions made by the robot that are desirable when operating under uncertainty. We show the benefit of this approach using simulations of a planar robot traversing an uncertain environment and of an underwater robot searching for loop-closure actions while performing visual SLAM.

[1]  Pieter Abbeel,et al.  Scaling up Gaussian Belief Space Planning Through Covariance-Free Trajectory Optimization and Automatic Differentiation , 2014, WAFR.

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

[3]  P. Abbeel,et al.  LQG-MP: Optimized path planning for robots with motion uncertainty and imperfect state information , 2011 .

[4]  Bruno Sinopoli,et al.  Kalman filtering with intermittent observations , 2004, IEEE Transactions on Automatic Control.

[5]  Nicholas Roy,et al.  Rapidly-exploring Random Belief Trees for motion planning under uncertainty , 2011, 2011 IEEE International Conference on Robotics and Automation.

[6]  Reid G. Simmons,et al.  Risk-Sensitive Planning with Probabilistic Decision Graphs , 1994, KR.

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

[8]  Wolfram Burgard,et al.  Information Gain-based Exploration Using Rao-Blackwellized Particle Filters , 2005, Robotics: Science and Systems.

[9]  Joelle Pineau,et al.  A Variance Analysis for POMDP Policy Evaluation , 2008, AAAI.

[10]  John N. Tsitsiklis,et al.  Bias and Variance Approximation in Value Function Estimates , 2007, Manag. Sci..

[11]  Héctor H. González-Baños,et al.  Navigation Strategies for Exploring Indoor Environments , 2002, Int. J. Robotics Res..

[12]  Jaime Valls Miró,et al.  Active Pose SLAM , 2012, 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems.

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

[14]  Pieter Abbeel,et al.  Risk Aversion in Markov Decision Processes via Near Optimal Chernoff Bounds , 2012, NIPS.

[15]  Daniel Hernández-Hernández,et al.  Risk Sensitive Markov Decision Processes , 1997 .

[16]  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..

[17]  Ryan M. Eustice,et al.  Real-Time Visual SLAM for Autonomous Underwater Hull Inspection Using Visual Saliency , 2013, IEEE Transactions on Robotics.

[18]  Jack Clark Francis,et al.  Modern portfolio theory : foundations, analysis, and new developments + website , 2013 .

[19]  Jerzy A. Filar,et al.  Variance-Penalized Markov Decision Processes , 1989, Math. Oper. Res..

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

[21]  Leslie Pack Kaelbling,et al.  Belief space planning assuming maximum likelihood observations , 2010, Robotics: Science and Systems.

[22]  Frank Dellaert,et al.  Square Root SAM: Simultaneous Localization and Mapping via Square Root Information Smoothing , 2006, Int. J. Robotics Res..

[23]  Geoffrey A. Hollinger,et al.  Sampling-based robotic information gathering algorithms , 2014, Int. J. Robotics Res..

[24]  Robert Buff.  Continuous Time Finance , 2002 .

[25]  Pieter Abbeel,et al.  LQG-MP: Optimized path planning for robots with motion uncertainty and imperfect state information , 2010, Int. J. Robotics Res..

[26]  Nicholas Roy,et al.  Global A-Optimal Robot Exploration in SLAM , 2005, Proceedings of the 2005 IEEE International Conference on Robotics and Automation.

[27]  M. J. Sobel.  The variance of discounted Markov decision processes , 1982 .