Zeta*-SIPP: Improved Time-Optimal Any-Angle Safe-Interval Path Planning
Authors: Yiyuan Zou, Clark Borst
IJCAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Benchmark experiments showed that Zeta*-SIPP reduced the computation time of TO-AA-SIPP by around 70%-90% on average. |
| Researcher Affiliation | Academia | Yiyuan Zou , Clark Borst Control and Simulation, Faculty of Aerospace Engineering Delft University of Technology, The Netherlands |
| Pseudocode | Yes | The pseudocode of Zeta-SIPP is shown in Algorithms 1-5. |
| Open Source Code | Yes | Our implementation is available at https://github.com/yiyuanzou/zeta-sipp. |
| Open Datasets | Yes | We performed experiments on three different benchmark maps [Stern et al., 2019]: Random-64-64-10, a 64 64 map with 10% of randomly blocked grids; Warehouse-10-20-102-2, a 170 84 map from a logistics domain; Berlin 1 256, a 256 256 real-world city map. |
| Dataset Splits | No | For each map, 500 scenarios were generated by the following steps: 1) Chose 25 benchmark scenario sets (random) [Stern et al., 2019]. 2) For each scenario set, we took the last 20 scenarios as tests and the top 32/64/96/128 scenarios as dynamic obstacles. The paper defines test scenarios but does not explicitly mention distinct training or validation splits in the context of model training. |
| Hardware Specification | Yes | All the algorithms were implemented in Java Script and the experiments were performed on Node.js v18.14.2 on a laptop with 2.30GHz Intel Core i7-11800H and 16 GB RAM. |
| Software Dependencies | Yes | All the algorithms were implemented in Java Script and the experiments were performed on Node.js v18.14.2 |
| Experiment Setup | Yes | For each map, 500 scenarios were generated by the following steps: 1) Chose 25 benchmark scenario sets (random) [Stern et al., 2019]. 2) For each scenario set, we took the last 20 scenarios as tests and the top 32/64/96/128 scenarios as dynamic obstacles. 3) The trajectories of dynamic obstacles were generated successively by Zeta*-SIPP, which were collision-free and contained any-angle moves. |