Planning under Uncertainty: Difference between revisions
(New page: In the Intelligent Systems Lab, one of the research foci is planning under uncertainty. That is, we compute plans for single agents as well as cooperative multiagent systems, in domains in...) |
|||
| (4 intermediate revisions by 2 users not shown) | |||
| Line 1: | Line 1: | ||
== Introduction == | |||
In the Intelligent Systems Lab, one of the research foci is planning under uncertainty. That is, we compute plans for single agents as well | In the Intelligent Systems Lab, one of the research foci is planning under uncertainty. That is, we compute plans for single agents as well | ||
as cooperative multiagent systems, in domains in which an agent is uncertain | as cooperative multiagent systems, in domains in which an agent is uncertain about the exact consequences of its actions. Furthermore, | ||
about the exact consequences of its actions. Furthermore, it is equipped with | it is equipped with imperfect sensors, resulting in noisy sensor readings which provide only limited | ||
imperfect sensors, resulting in noisy sensor readings which provide only limited | information. | ||
information. For single agents, such planning problems are naturally framed in | |||
For single agents, such planning problems are naturally framed in | |||
the partially observable Markov decision process (POMDP) paradigm. In a POMDP, uncertainty in acting | the partially observable Markov decision process (POMDP) paradigm. In a POMDP, uncertainty in acting | ||
and sensing is captured in probabilistic models, and allows an agent to plan | and sensing is captured in probabilistic models, and allows an agent to plan | ||
| Line 10: | Line 12: | ||
the decentralized POMDP (Dec-POMDP) framework. | the decentralized POMDP (Dec-POMDP) framework. | ||
== Projects == | |||
* [http://gaips.inesc-id.pt/inside/ INSIDE] | |||
* [http://gaips.inesc-id.pt/mais-s/ IMAIS+S] | |||
* [http://decpucs.isr.ist.utl.pt DecPUCS] | |||
== Software == | |||
* [http://wiki.ros.org/markov_decision_making Markov Decision Making package for ROS], by João Messias | |||
* [http://www.fransoliehoek.net/index.php?fuseaction=software.madp MADP], Multiagent decision process toolbox by Matthijs Spaan and Frans Oliehoek | |||
* [http://users.isr.ist.utl.pt/~mtjspaan/decpomdp Dec-POMDP problem domains], by Matthijs Spaan | |||
* [http://dante.isr.tecnico.ulisboa.pt/tsveiga/POMDPSolvers POMDP Solvers], repository created by Tiago Veiga - check the folder | |||
'SymbolicPerseusIR' for Tiago's POMDP-IR code, that endows POMDPs with special actions which enable rewarding beliefs instead | |||
of states, without modifying the traditional solution of PODMPs (value function remains PWLC). | |||
Check also | |||
* [http://www.openmarkov.org/ OpenMarkov] - output .pgmx files are recognized by Tiago's POMDP-IR code and MADP | |||
Latest revision as of 16:23, 19 November 2016
Introduction
In the Intelligent Systems Lab, one of the research foci is planning under uncertainty. That is, we compute plans for single agents as well as cooperative multiagent systems, in domains in which an agent is uncertain about the exact consequences of its actions. Furthermore, it is equipped with imperfect sensors, resulting in noisy sensor readings which provide only limited information.
For single agents, such planning problems are naturally framed in the partially observable Markov decision process (POMDP) paradigm. In a POMDP, uncertainty in acting and sensing is captured in probabilistic models, and allows an agent to plan on its belief state, which summarizes all the information the agent has received regarding its environment. For the multiagent case, we frame our planning problem in the decentralized POMDP (Dec-POMDP) framework.
Projects
Software
- Markov Decision Making package for ROS, by João Messias
- MADP, Multiagent decision process toolbox by Matthijs Spaan and Frans Oliehoek
- Dec-POMDP problem domains, by Matthijs Spaan
- POMDP Solvers, repository created by Tiago Veiga - check the folder
'SymbolicPerseusIR' for Tiago's POMDP-IR code, that endows POMDPs with special actions which enable rewarding beliefs instead of states, without modifying the traditional solution of PODMPs (value function remains PWLC).
Check also
- OpenMarkov - output .pgmx files are recognized by Tiago's POMDP-IR code and MADP