Importance Sampling for Online Planning under Uncertainty
نویسندگان
چکیده
The partially observable Markov decision process (POMDP) provides a principled general framework for robot planning under uncertainty. Leveraging the idea of Monte Carlo sampling, recent POMDP planning algorithms have scaled up to various challenging robotic tasks, including, e.g., real-time online planning for autonomous vehicles. To further improve online planning performance, this paper presents IS-DESPOT, which introduces importance sampling to DESPOT, a state-of-the-art sampling-based POMDP algorithm for planning under uncertainty. Importance sampling improves the planning performance when there are critical, but rare events, which are difficult to sample. We prove that IS-DESPOT retains the theoretical guarantee of DESPOT. We present a general method for learning the importance sampling distribution and demonstrate empirically that importance sampling significantly improves the performance of online POMDP planning for suitable tasks.
منابع مشابه
Scalable Multiagent Coordination with Distributed Online Open Loop Planning
We propose distributed online open loop planning (DOOLP), a general framework for online multiagent coordination and decision making under uncertainty. DOOLP is based on online heuristic search in the space defined by a generative model of the domain dynamics, which is exploited by agents to simulate and evaluate the consequences of their potential choices. We also propose distributed online Th...
متن کاملOn the Probabilistic Completeness of the Sampling-based Motion Planning Methods Under Uncertainty
This paper extends the concept of probabilistic completeness defined for the motion planners in the absence of noise, to the concept of “probabilistic completeness under uncertainty” for the motion planners that perform planning in the presence of uncertainty. According to the proposed definition, an approach is proposed to verify the probabilistic completeness under uncertainty. Finally, it is...
متن کاملMonte Carlo Motion Planning for Robot Trajectory Optimization Under Uncertainty
This article presents a novel approach, named MCMP (Monte Carlo Motion Planning), to the problem of motion planning under uncertainty, i.e., to the problem of computing a low-cost path that fulfills probabilistic collision avoidance constraints. MCMP estimates the collision probability (CP) of a given path by sampling via Monte Carlo the execution of a reference tracking controller (in this pap...
متن کاملSimultaneous Localization and Planning for Physical Mobile Robots via Enabling Dynamic Replanning in Belief Space
Simultaneous planning while localizing is a crucial ability for an autonomous robot operating under uncertainty. This paper addresses this problem by designing methods to dynamically replan while the localization uncertainty or environment map is updated. In particular, relying on sampling-based methods, the proposed online planning scheme can cope with challenging situations, including when th...
متن کاملA Domain Decomposition Approach for Uncertainty Analysis
This paper proposes a decomposition approach for uncertainty analysis of systems governed by partial differential equations (PDEs). The system is split into local components using domain decomposition. Our domain-decomposed uncertainty quantification (DDUQ) approach performs uncertainty analysis independently on each local component in an “offline” phase, and then assembles global uncertainty a...
متن کامل