Dynamically Constructed (PO)MDPs for Adaptive Robot Planning

Authors: Shiqi Zhang, Piyush Khandelwal, Peter Stone

AAAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We conduct a number of experimental trials using two example problems in simulation and demonstrate i CORPP on a real robot. Results show significant improvements compared to competitive baselines.
Researcher Affiliation Academia Shiqi Zhang,1,2 Piyush Khandelwal,2 Peter Stone2 1 Department of Electrical Engineering and Computer Science, Cleveland State University 2 Department of Computer Science, The University of Texas at Austin s.zhang9@csuohio.edu; {piyushk,pstone}@cs.utexas.edu
Pseudocode Yes Algorithm 1 specifies the i CORPP algorithm.
Open Source Code No The paper does not provide concrete access to source code for the methodology described in this paper.
Open Datasets No The paper describes tasks within a simulation environment (GAZEBO) and on a real robot, but does not explicitly mention or provide access information for a publicly available or open dataset used for training/evaluation.
Dataset Splits No The paper does not provide specific dataset split information (percentages, sample counts, or citations to predefined splits) needed to reproduce the data partitioning.
Hardware Specification No The paper mentions demonstration "on a real robot" and "simulation environment", but does not provide specific hardware details (e.g., CPU/GPU models, memory specifications) used for running experiments.
Software Dependencies No We used a solver introduced in (Zhu 2012) for P-LOG programs (except that reasoning about reward was manually conducted), the APPL solver for POMDPs (Kurniawati, Hsu, and Lee 2008), and value iteration for MDPs (Sutton and Barto 1998).
Experiment Setup Yes In the default and cautious versions of i CORPP, the values of [R+,R ] are [20, 20] and [30, 30] respectively. ... We limit the number of random walkers to be 1 and its speed to be one fifth of the robot s.