Parametric POMDPs for planning in continuous state spaces

Brooks, Alex; Makarenko, Alexei; Williams, Stefan B.; Durrant‐Whyte, Hugh

doi:10.1016/j.robot.2006.05.007

Cited by 73 publications

(58 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A belief compression technique that assumes a unimodal belief state, such as Brooks et al [2] approach, cannot perform optimally in this domain. This is because the belief state resulting from an action is inherently bimodal due to the faulty dynamics model, and a unimodal approach must, for example, average, or select one peak of the belief.…”

Section: Locomotion Over Rough Terrain: Bimodal Dynamicsmentioning

confidence: 99%

“…Brooks et al [2] restrict the belief state to be a unimodal Gaussian, 1 and then plan by discretizing the Gaussian parameters and using Fitted Value Iteration over that representation. However, this approach will be limited in any domain that requires multi-modal beliefs to achieve good performance.…”

Section: Introductionmentioning

confidence: 99%

“…The work by Brooks [2], Porta [15] and their colleagues presented results on problems that used the simple linear Gaussian to describe the effects of an action. However, the dynamics of robotic grasping, autonomous navigation over varying terrain, and many other problems of interest are highly complex and nonlinear.…”

Section: Introductionmentioning

confidence: 99%

“…In contrast to most 2 prior approximate continuous-state POMDP work [2,15,21], we also provide a theoretical analysis of several variants of our SM-POMDP planner, bounding where possible the error between the resulting SM-POMDP value function, and the optimal value function. We also analyze the computational complexity of planning using SM-POMDP.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Planning in partially-observable switching-mode continuous domains

Brunskill

Kaelbling

Lozano-Pérez

et al. 2010

Ann Math Artif Intell

View full text Add to dashboard Cite

Continuous-state POMDPs provide a natural representation for a variety of tasks, including many in robotics. However, most existing parametric continuousstate POMDP approaches are limited by their reliance on a single linear model to represent the world dynamics. We introduce a new switching-state dynamics model that can represent multi-modal state-dependent dynamics. We present the Switching Mode POMDP (SM-POMDP) planning algorithm for solving continuousstate POMDPs using this dynamics model. We also consider several procedures to approximate the value function as a mixture of a bounded number of Gaussians. Unlike the majority of prior work on approximate continuous-state POMDP planners, we provide a formal analysis of our SM-POMDP algorithm, providing bounds, where possible, on the quality of the resulting solution. We also analyze the computational complexity of SM-POMDP. Empirical results on an unmanned aerial vehicle collisions avoidance simulation, and a robot navigation simulation where the robot has faulty actuators, demonstrate the benefit of SM-POMDP over a prior parametric approach.

show abstract

Section: Locomotion Over Rough Terrain: Bimodal Dynamicsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Planning in partially-observable switching-mode continuous domains

Brunskill

Kaelbling

Lozano-Pérez

et al. 2010

Ann Math Artif Intell

View full text Add to dashboard Cite

show abstract

“…Instead, moving vehicles are often assumed to follow known trajectories or policies, which the planner reacts to locally. While full probabilistic representation of the environment is possible, such as with POMDPs [2], [3], the corresponding solution techniques are computationally intractible for real-time path planning problems of even modest complexity or dimension. Recently, sampling-based techniques such as rapidlyexploring random trees (RRTs) [4] have been extended to incorporate motion patterns for moving obstacles learned offline via Gaussian processes [5].…”

Section: Introductionmentioning

confidence: 99%

Threat-aware path planning in uncertain urban environments

Aoude

Luders

Levine

et al. 2010

2010 IEEE/RSJ International Conference on Intelligent Robots and Systems

View full text Add to dashboard Cite

Abstract-This paper considers the path planning problem for an autonomous vehicle in an urban environment populated with static obstacles and moving vehicles with uncertain intents. We propose a novel threat assessment module, consisting of an intention predictor and a threat assessor, which augments the host vehicle's path planner with a real-time threat value representing the risks posed by the estimated intentions of other vehicles. This new threat-aware planning approach is applied to the CL-RRT path planning framework, used by the MIT team in the 2007 DARPA Grand Challenge. The strengths of this approach are demonstrated through simulation and experiments performed in the RAVEN testbed facilities.

show abstract

Decision Making Under Uncertainty

Sebbane

2014

Intelligent Systems, Control and Automation: Science and Engineering

View full text Add to dashboard Cite

Parametric POMDPs for planning in continuous state spaces

Cited by 73 publications

References 18 publications

Planning in partially-observable switching-mode continuous domains

Planning in partially-observable switching-mode continuous domains

Threat-aware path planning in uncertain urban environments

Decision Making Under Uncertainty

Contact Info

Product

Resources

About