Intelligent Agent Supporting Human-Multi-Robot Team Collaboration
Authors: Ariel Rosenfeld, Noa Agmon, Oleg Maksimov, Amos Azaria, Sarit Kraus
IJCAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our intelligent advising agent was evaluated through extensive field trials, with 44 non-expert human operators and 10 low-cost mobile robots, in simulation and physical deployment, and showed a significant improvement in both team performance and the operator s satisfaction. |
| Researcher Affiliation | Academia | 1 Department of Computer Science, Bar-Ilan University, Ramat-Gan, Israel 52900. 2 Machine Learning Department, Carnegie Mellon University, Pittsburgh, USA. |
| Pseudocode | Yes | Algorithm 1 Advice Provision |
| Open Source Code | No | The paper does not provide concrete access to source code for the methodology described. It only provides a link to a video of the project: 'A short video of the project is available at http://vimeo.com/119434903.' |
| Open Datasets | No | The paper describes specific environments (office building, Counter Strike map) and the generation of data through simulations and physical deployment for training models, but it does not provide concrete access information (specific link, DOI, repository name, or formal citation with authors/year) for a publicly available or open dataset. |
| Dataset Splits | No | The paper describes a within-subject experimental design with human operators and details about the experimental conditions (e.g., malfunction schedules, object positions) and training of the agent's models using simulation sessions. However, it does not provide specific train/validation/test dataset splits (exact percentages, sample counts, or citations to predefined splits) for a static dataset that would be needed for direct reproduction of data partitioning in a machine learning context. |
| Hardware Specification | Yes | We used 10 Hamster AUGVs (autonomous unmanned ground vehicles) (See Figure 2). Hamster is a 4WD rugged platform with a built-in navigation algorithm that allows it to explore, map and localize in unknown areas. Hamster has 2 on-board Raspberry PI Linux servers for algorithm execution and an Arduino for low level control. Hamster is mounted with an HD camera with h264 video streaming over Wi Fi and a 360 6-meter range LIDAR laser. Each Hamster is 190mm in width, 240mm in length and 150mm in height. |
| Software Dependencies | No | The paper mentions using the 'Gazebo robot simulation toolbox' and 'ROS', but does not provide specific version numbers for these software dependencies or any other ancillary software components, which is required for reproducible description. |
| Experiment Setup | Yes | To learn R( ), we ran 150 1-hour sessions on each environment. During these sessions, the robots autonomously searched for 40 randomly placed green objects. In each session we placed random obstacles and used a different number of robots ranging from 1 to 10. ... The SAR task took 40 minutes (in simulation) and 15 minutes (in physical deployment)... In Environment 1s (simulated office building), we placed 40 green objects around the office. We set a malfunction schedule such that each robot encounters 1.5 malfunctions (on average) during the simulation. |