Grounding State Representations in Sensory Experience for Reasoning and Planning by Mobile Robots

Abstract

We are addressing the problem of learning probabilis-tic models of the interaction between a mobile robot and its environment and using these models for task planning. This requires modifying the state-of-the-art reinforcement learning algorithms to deal with hidden state and high-dimensional observation spaces of con-tinuous variables. Our approach is to identify hidden states by means of the trajectories leading into and out of them, and perform clustering in this embedding tra-jectory space in order to compile a partially observable Markov decision process (POMDP) model, which can be used for approximate decision-theoretic planning. The ultimate objective of our work is to develop algo-rithms that learn POMDP models with discrete hidden

Cite

Text

Nikovski. "Grounding State Representations in Sensory Experience for Reasoning and Planning by Mobile Robots." AAAI Conference on Artificial Intelligence, 2000.

Markdown

[Nikovski. "Grounding State Representations in Sensory Experience for Reasoning and Planning by Mobile Robots." AAAI Conference on Artificial Intelligence, 2000.](https://mlanthology.org/aaai/2000/nikovski2000aaai-grounding/)

BibTeX

@inproceedings{nikovski2000aaai-grounding,
  title     = {{Grounding State Representations in Sensory Experience for Reasoning and Planning by Mobile Robots}},
  author    = {Nikovski, Daniel},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2000},
  pages     = {1108},
  url       = {https://mlanthology.org/aaai/2000/nikovski2000aaai-grounding/}
}