Task Planning for Object Rearrangement in Multi-Room Environments
Authors: Karan Mirakhor, Sourav Ghosh, Dipanjan Das, Brojeshwar Bhowmick
AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The paper also presents new metrics and a benchmark dataset called Mo POR to evaluate the effectiveness of the rearrangement planning in a multi-room setting. The experimental results demonstrate that the proposed method effectively addresses the multi-room rearrangement problem. |
| Researcher Affiliation | Industry | Visual Computing and Embodied Intelligence Lab, TCS Research, Kolkata, India {karan.mirakhor, g.sourav10, dipanjan.da, b.bhowick}@tcs.com |
| Pseudocode | Yes | Algorithm 1: Algorithm for Task planner |
| Open Source Code | Yes | Code link : https://tinyurl.com/MultiRoomCode |
| Open Datasets | Yes | Unseen Object Discovery Dataset The AMT dataset (Kant et al. 2022) consists of 268 object categories present in 12 distinct rooms and 32 receptacle types. ... We present the Mo POR Benchmark Dataset to overcome the shortcomings in existing benchmark (Weihs et al. 2021) for assessing multi-room task planning. |
| Dataset Splits | No | The paper states, 'Training The training details of UODM and DSG with Deep RL planner are available in the Appendix2.' However, it does not explicitly specify dataset split percentages or counts for training, validation, or test sets in the main text. |
| Hardware Specification | No | The paper does not provide specific details about the hardware (e.g., GPU/CPU models, memory) used for running the experiments in the main text. |
| Software Dependencies | No | The paper mentions using specific tools like Ai2Thor and d-DETR, but does not provide version numbers for any software dependencies. |
| Experiment Setup | No | The paper states: 'Training The training details of UODM and DSG with Deep RL planner are available in the Appendix2.' This indicates that detailed experimental setup information, such as hyperparameters or specific training configurations, is not present in the main text. |