Touch based POMDP manipulation via sequential submodular optimization

Vien Ngo, Marc Toussaint

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)


Exploiting the submodularity of entropy-related objectives has recently led to a series of successes in machine learning and sequential decision making. Its generalized framework, adaptive submodularity, has later been introduced to deal with uncertainty and partially observability, achieving near-optimal performance with simple greedy policies. As a consequence, adaptive submodularity is in principle a promising candidate for efficient touch-based localization in robotics. However, applying that method directly on the motion level shows poor scaling with the dimensionality of the system. Being motivated by hierarchical partially observable Markov decision process (POMDP) planning, we integrate an action hierarchy into the existing adaptive submodularity framework. The proposed algorithm is expected to effectively generate uncertainty-reducing actions with the help from an action hierarchy. Experimental results on both, a simulated robot and a Willow Garage PR2 platform, demonstrate the efficiency of our algorithm.
Original languageEnglish
Title of host publication15th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2015, Seoul, South Korea, November 3-5, 2015
Subtitle of host publicationHumanoids
Publisher IEEE
Number of pages7
ISBN (Electronic)978-1-4799-6885-5
Publication statusPublished - 28 Dec 2015


Dive into the research topics of 'Touch based POMDP manipulation via sequential submodular optimization'. Together they form a unique fingerprint.

Cite this