Qualitative Planning with Quantitative Constraints for Online Learning of Robotic Behaviours

Authors: Timothy Wiley, Claude Sammut, Ivan Bratko

AAAI 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental To evaluate the impact of quantitative constraints, experiments were conducted that compared the performance of the existing qualitative-only planner to the planner extended with quantitative constraints. The experiments were conducted using the step climbing task.
Researcher Affiliation Academia Timothy Wiley and Claude Sammut School of Computer Science and Engineering University of New South Wales Sydney, NSW 2052, Australia {timothyw,claude}@cse.unsw.edu.au Ivan Bratko Faculty of Computer and Information Science University of Ljubljana Trzaska 25, 1000 Ljubljana, Slovenia bratko@fri.uni-lj.si
Pseudocode No The paper does not contain any pseudocode or clearly labeled algorithm blocks.
Open Source Code No The paper mentions a 'Prolog implementation' of the qualitative planner and QSIM, and refers to 'ASPQSIM' as a more efficient implementation (Wiley, Sammut, and Bratko 2014), but it does not explicitly state that the code for the methodology described in this paper is open-source or publicly released.
Open Datasets No The paper describes a specific application domain (step climbing for the iRobot Negotiator) but does not provide access information (link, DOI, citation) to a publicly available or open dataset used for training. The environment and task are described conceptually.
Dataset Splits No The paper does not specify exact dataset split percentages, absolute sample counts, or reference predefined splits for training, validation, or testing.
Hardware Specification No The paper states that 'The performance of the planner is compared by the number of inferences Prolog evaluates to find a plan as it is independent of variations in a specific CPU,' but it does not specify any particular CPU model, GPU, or other hardware used for running the experiments.
Software Dependencies No The paper mentions 'a Prolog implementation of the qualitative planner, QSIM (Bratko 2011), and A* (Hart, Nilsson, and Raphael 1968),' but it does not provide specific version numbers for Prolog, QSIM, or any other ancillary software components used in the experiments.
Experiment Setup Yes For each experiment, the robot was initially stationary, on the ground before the step, with the flippers directed toward the step, (posx = 0, hd = 0, v = 0). The experiments are grouped by the variables specified in the goal state. For each goal state, the choice of heuristic and the type of planner (with quantitative constraints optionally enabled) was varied. In the first set of experiments, the step height was set to 10 cm... Finally, for reference a set of experiments was conducted with a step height of 30 cm.