Maximum Sustainable Yield Problem for Robot Foraging and Construction System
Authors: Ruohan Zhang, Zhao Song
IJCAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | 5 Experiments and Results We compare the performance of the informed algorithm and the adaptive algorithm. We implement a simulator using Player/Stage [Gerkey et al., 2003]. The parameters are chosen to be consistent with [Song and Vaughan, 2013]. |
| Researcher Affiliation | Academia | Ruohan Zhang and Zhao Song Department of Computer Science, University of Texas at Austin, Austin, United States {zharu,zhaos}@utexas.edu |
| Pseudocode | Yes | Algorithm 1 Construction Robot Algorithm (Informed) ... Algorithm 2 Foraging Robot Algorithm (Adaptive) |
| Open Source Code | No | The paper does not provide any concrete access information (link, explicit statement of release) to the source code for the methodology described. |
| Open Datasets | No | The paper describes a simulated environment with specified parameters for resource growth and robot behavior, but it does not use or provide access to a publicly available or open dataset. |
| Dataset Splits | No | The paper describes simulation parameters and compares algorithms within a simulated environment, but it does not define specific training, validation, or test dataset splits as typically found in data-driven machine learning research. |
| Hardware Specification | No | The paper states 'We implement a simulator using Player/Stage [Gerkey et al., 2003]' but does not provide any specific details about the hardware (e.g., CPU, GPU, memory) used to run these simulations or experiments. |
| Software Dependencies | No | The paper mentions 'We implement a simulator using Player/Stage [Gerkey et al., 2003]', but it does not provide specific version numbers for Player/Stage or any other software dependencies. |
| Experiment Setup | Yes | The parameters are chosen to be consistent with [Song and Vaughan, 2013]. Resources: the number of patches |A| = 3. We use the same logistic model parameters for all patches, k = .4, b = .004. ... The lower bound, upper bound, and initial population are P0 = 15, P1 = 100, P(0) = 50, respectively. The resource growth time interval s = 100. ... Construction robot: the construction time for a new robot tconstruct = 1200. The amount of pucks needed to construct a new robot w = [3, 2, 1]. Foraging robots: tharvest = 10, tunload = 8, velocity = .004. ... The parameters we use for our algorithms are Kin = 3.0, Kde = 2.0 for the informed sleep time adjustment. Kin = 3.0, Kde = .2 for the adaptive sleep time adjustment. ... We use a Gaussian filter of σ = 10 with kernel size 20 to smooth the observed data. The total length of experiment is 20, 0000 time steps and we record data every 200 steps. The noises are Gaussian random variables with mean zero and standard deviations equal 10% of the affected variable (Equation 12). |