Improving Surveillance Using Cooperative Target Observation
Authors: Rashi Aswani, Sai Krishna Munnangi, Praveen Paruchuri
AAAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We then modify the observer strategy proposed in the literature based on the K-means algorithm to introduce five variants and provide experimental validation. Each experiment was simulated 30 times (number of runs) and each simulation run consists of a total of 1500 time-steps. |
| Researcher Affiliation | Academia | Rashi Aswani, Sai Krishna Munnangi and Praveen Paruchuri Machine Learning Lab, Kohli Center on Intelligent Systems International Institute of Information Technology Hyderabad, India {rashi.aswani, krishna.munnangi}@research.iiit.ac.in, praveen.p@iiit.ac.in |
| Pseudocode | Yes | Algorithm 1 BRLP-CTO(Emin, p (α)) 1: Set βl = 0, βu = 1 and β = 1/2. 2: Solve Problem (3), let p(α)β and E(β) be the optimal solution and expected reward value obtained. 3: if Emin > E then 4: while |E(β) Emin| > ϵ do 5: if E(β) > Emin then 6: Set βl = β 7: else 8: Set βu = β 9: end if 10: β = βl+βu 2 11: Solve Problem (3), let p(α)β and E(β) be the optimal solution and expected reward value returned 12: end while 13: end if 14: return p(α)β |
| Open Source Code | No | The paper does not provide explicit statements or links indicating that the source code for their methodology is open-source or publicly available. |
| Open Datasets | No | For purposes of experiments, we assume that the observers and targets are operating in a rectangular field with a width and height of 150 x 150 units (we refer to as units for generalization). The paper describes a simulation environment and its parameters rather than using or providing a link to a publicly available dataset. |
| Dataset Splits | No | The paper describes running simulations under various parameter settings, such as varying density, speed, and sensor range for agents, but does not specify traditional training, validation, or test dataset splits. |
| Hardware Specification | No | The paper mentions using the MASON simulation toolkit and its internal threading mechanism, but it does not provide any specific details about the hardware (e.g., CPU, GPU models, memory) used for running the experiments. |
| Software Dependencies | No | All our experiments were performed on MASON simulation toolkit (Luke et al. 2004). MASON is a fast discrete-event multi-agent simulation library core developed in Java. (It mentions MASON and Java but no version numbers). |
| Experiment Setup | Yes | For purposes of experiments, we assume that the observers and targets are operating in a rectangular field with a width and height of 150 x 150 units. To vary density, the number of observers was picked from {2, 6, 10, 14, 18} and the number of targets from {3, 9, 15, 21, 27}. Six possible target speed values were picked among {0.2, 0.5, 0.8, 1.0, 1.2, 1.5} measured as units per timestep. The speed of observers was fixed to 1 unit per timestep. Five possible sensor range values picked from {5, 10, 15, 20, 25}. |