Grandpa Hates Robots – Interaction Constraints for Planning in Inhabited Environments

Authors: Uwe Koeckemann, Federico Pecora, Lars Karlsson

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

Reproducibility Variable Result LLM Response
Research Type Experimental We illustrate the computational advantage this brings with experiments on a large-scale (semi-)realistic household domain with hundreds of human activities and several robots. Experimental Setup. The purpose of the experiments is to show that the proposed planner is capable of creating plans that satisfy a set of ICs for problem instances of a reasonable size. Experimental Results.
Researcher Affiliation Academia Uwe K ockemann and Federico Pecora and Lars Karlsson Center for Applied Autonomous Sensor Systems, Orebro University, Sweden {uwe.koeckemann,federico.pecora,lars.karlsson}@oru.se
Pseudocode Yes Algorithm 1 Constraint-based planning; Algorithm 2 Resolving interaction constraints
Open Source Code No The paper does not provide any explicit statements about releasing source code or links to a code repository for the methodology described.
Open Datasets No We randomly generated complete human schedules in the following way: Starting at 6:00am, every activity lasts between 30 and 60 minutes. After filling a 16 hour time frame in this manner, the activity of every family member is set to sleep for the rest of the day. The paper does not provide concrete access information (link, DOI, citation) for a publicly available or open dataset.
Dataset Splits No The paper mentions generating human schedules and varying parameters like the number of inhabitants and goals. It does not provide specific details on training, validation, and test dataset splits for reproducibility.
Hardware Specification Yes Experiments were run on an Intel R Core TM i7-2620M CPU @ 2.70GHz x 4, 4GB RAM platform with a 3GB memory limit.
Software Dependencies No The paper mentions using Prolog and references external planning systems, but it does not specify version numbers for any software dependencies used in their own implementation.
Experiment Setup Yes We randomly generated complete human schedules in the following way: Starting at 6:00am, every activity lasts between 30 and 60 minutes. We used three different sets of humans h1, h2 and h3 containing 5, 7 and 12 inhabitants. The number of goals varies between 10 and 30 (g10 and g30) and goals are randomly distributed into the three goal batches. We used two configurations of uncertainty u0 (no uncertainty) and u1 (with uncertainty). For each activity created with u1 there is a 10% chance that this activity will be only partially specified and we randomly pick between two and four random possible values. We set a time limit of 30 minutes to solve each problem.