Integrated perception and planning in the continuous space: A POMDP approach

Bai, Haoyu; Hsu, David; Lee, Wee Sun

doi:10.1177/0278364914528255

Cited by 101 publications

(95 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finite-state controllers explicitly represent infinite-horizon policies, but can also be used (as a possibly more concise representation) for finite horizon policies. They have been widely used in POMDPs 1 (e.g., see Kaelbling et al 1998, Hansen 1998, Meuleau et al 1999b, Poupart and Boutilier 2004, Poupart 2005, Toussaint et al 2006, 2008, Grześ et al 2013, Bai et al 2014 and Dec-POMDPs (e.g., Bernstein et al 2005, Amato et al 2007a, Bernstein et al 2009, Kumar and Zilberstein 2010b, Pajarinen and Peltonen 2011a, Kumar et al 2011, Pajarinen and Peltonen 2011b, Wu et al 2013.…”

Section: Policy Representationmentioning

confidence: 99%

A Concise Introduction to Decentralized POMDPs

Oliehoek

Amato

2016

SpringerBriefs in Intelligent Systems

524

233

View full text Add to dashboard Cite

Section: Policy Representationmentioning

confidence: 99%

A Concise Introduction to Decentralized POMDPs

Oliehoek

Amato

2016

SpringerBriefs in Intelligent Systems

524

233

View full text Add to dashboard Cite

“…For clarity of presentation, however, we will proceed with the entire joint state X k+l which contains X c k+l . 1 We thus define the generalized belief space (GBS) at the lth planning step as…”

Section: Approach Overviewmentioning

confidence: 99%

“…The approaches falling in these second category usually assumes that the robot moves in a known environment; a remarkable property of these techniques is that they approach optimality when increasing the runtime (which is exponential in the size of the problem). A recent example of infinite-horizon planning is the work [1], in which Bai et al apply a Monte Carlo sampling technique to update an initial policy, assuming maximum likelihood observations. Finally, receding horizon strategies compute a policy over the next L control actions, where L is a given horizon.…”

Section: Introductionmentioning

confidence: 99%

Towards Planning in Generalized Belief Space

Indelman

Carlone

Dellaert

2016

Springer Tracts in Advanced Robotics

View full text Add to dashboard Cite

We investigate the problem of planning under uncertainty, which is of interest in several robotic applications, ranging from autonomous navigation to manipulation. Recent effort from the research community has been devoted to design planning approaches working in a continuous domain, relaxing the assumption that the controls belong to a finite set. In this case robot policy is computed from the current robot belief (planning in belief space), while the environment in which the robot moves is usually assumed to be known or partially known. We contribute to this branch of the literature by relaxing the assumption of known environment; for this purpose we introduce the concept of generalized belief space (GBS), in which the robot maintains a joint belief over its state and the state of the environment. We use GBS within a Model Predictive Control (MPC) scheme; our formulation is valid for general cost functions and incorporates a dual-layer optimization: the outer layer computes the best control action, while the inner layer computes the generalized belief given the action. The resulting approach does not require prior knowledge of the environment and does not assume maximum likelihood observations. We also present an application to a specific family of cost functions and we elucidate on the theoretical derivation with numerical examples.

show abstract

“…2) Environment with Beacons-We also consider a scenario where the car-like robot estimates its location using measurements from two beacons, b 1 6 shows the B-SELQR trajectory and associated beliefs along the trajectory. The computed control policies steer the robot to move in the vicinity of the two beacons in order to better localize the robot before proceeding to the goal.…”

Section: ) Light-dark Environment-mentioning

confidence: 99%

Stochastic Extended LQR: Optimization-Based Motion Planning Under Uncertainty

Sun

Berg

Alterovitz

2015

Springer Tracts in Advanced Robotics

View full text Add to dashboard Cite

We introduce a novel optimization-based motion planner, Stochastic Extended LQR (SELQR), which computes a trajectory and associated linear control policy with the objective of minimizing the expected value of a user-defined cost function. SELQR applies to robotic systems that have stochastic non-linear dynamics with motion uncertainty modeled by Gaussian distributions that can be state-and control-dependent. In each iteration, SELQR uses a combination of forward and backward value iteration to estimate the cost-to-come and the cost-to-go for each state along a trajectory. SELQR then locally optimizes each state along the trajectory at each iteration to minimize the expected total cost, which results in smoothed states that are used for dynamics linearization and cost function quadratization. SELQR progressively improves the approximation of the expected total cost, resulting in higher quality plans. For applications with imperfect sensing, we extend SELQR to plan in the robot's belief space. We show that our iterative approach achieves fast and reliable convergence to high-quality plans in multiple simulated scenarios involving a car-like robot, a quadrotor, and a medical steerable needle performing a liver biopsy procedure.

show abstract

Integrated perception and planning in the continuous space: A POMDP approach

Cited by 101 publications

References 16 publications

A Concise Introduction to Decentralized POMDPs

A Concise Introduction to Decentralized POMDPs

Towards Planning in Generalized Belief Space

Stochastic Extended LQR: Optimization-Based Motion Planning Under Uncertainty

Contact Info

Product

Resources

About