MMUSV-Sim: A Maritime Simulation Platform for Multi-USV Cooperative Perception
MMUSV-Sim: A Maritime Simulation Platform for Multi-USV Cooperative Perception
This paper presents MMUSV-Sim, a maritime simulation platform built on Unreal Engine 5 and Project AirSim designed to enable cooperative perception among multiple unmanned surface vehicles (USVs). The authors address challenges such as limited sensing range, occlusions, and the high cost of sea trials by providing a configurable environment for synchronized multimodal data acquisition. Using this platform, they generated the MarU2U-200m dataset, comprising 341 sequences and 14,385 scene frames with consistent 3D annotations. The study establishes a cooperative bird's-eye-view vessel detection benchmark, demonstrating that Early Fusion improves LiDAR detection [email protected] from 64.26 to 81.46 compared to single-USV baselines, while RGB features provide further improvements. The work also evaluates the utility of generated data for pretraining real-world maritime detection models.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- Environment
- SimulationOpen source S3
Reported numbers
scene frames
14,385
View original evidence
Using MMUSV-Sim, we construct MarU2U-200m, a synchronized multi-USV RGB--LiDAR dataset with 341 sequences and 14,385 scene frames. We establish a cooperative bird's-eye-view (BEV) vessel detection benchmark with LiDAR and RGB--LiDAR inputs. Experimental results show that Early Fusion improves LiDAR detection [email protected] from 64.26 to 81.46 compared with the single-USV baseline. RGB features further improve intermediate f
Open source S4sequences
341
View original evidence
Using MMUSV-Sim, we construct MarU2U-200m, a synchronized multi-USV RGB--LiDAR dataset with 341 sequences and 14,385 scene frames. We establish a cooperative bird's-eye-view (BEV) vessel detection benchmark with LiDAR and RGB--LiDAR inputs. Experimental results show that Early Fusion improves LiDAR detection [email protected] from 64.26 to 81.46 compared with the single-USV baseline. RGB features further improve intermediate f
Open source S4AP
64.26–81.46
View original evidence
Using MMUSV-Sim, we construct MarU2U-200m, a synchronized multi-USV RGB--LiDAR dataset with 341 sequences and 14,385 scene frames. We establish a cooperative bird's-eye-view (BEV) vessel detection benchmark with LiDAR and RGB--LiDAR inputs. Experimental results show that Early Fusion improves LiDAR detection [email protected] from 64.26 to 81.46 compared with the single-USV baseline. RGB features further improve intermediate f
Open source S4LiDAR detection AP
64.26–81.46
View original evidence
Using MMUSV-Sim, we construct MarU2U-200m, a synchronized multi-USV RGB--LiDAR dataset with 341 sequences and 14,385 scene frames. We establish a cooperative bird's-eye-view (BEV) vessel detection benchmark with LiDAR and RGB--LiDAR inputs. Experimental results show that Early Fusion improves LiDAR detection [email protected] from 64.26 to 81.46 compared with the single-USV baseline. RGB features further improve intermediate f
Open source S4
Dataset details reported by the source
- Modalities
- RGB Open source S4LiDAR Open source S4
- Tasks
- vessel detection Open source S4
- Embodiments
- multi-USV Open source S4
- dataset: not_reported
- paper: not_reported
- weights: not_reported
- code: not_reported
What remains unknown
Dataset licensing and commercial-use rights are not established here.
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
dinated sea trials, the scarcity of multi-USV datasets, and the lack of integrated maritime simulation and evaluation tools. We present MMUSV-Sim, a perception-oriented maritime simulation and data-generation platform built on Unreal Engine 5 and Project AirSim. It supports configurable maritime scenarios, synchronized multimodal acquisition across multiple USVs, and consistent 3D annotations in a shared world frame.
Open source S3
Using MMUSV-Sim, we construct MarU2U-200m, a synchronized multi-USV RGB--LiDAR dataset with 341 sequences and 14,385 scene frames. We establish a cooperative bird's-eye-view (BEV) vessel detection benchmark with LiDAR and RGB--LiDAR inputs. Experimental results show that Early Fusion improves LiDAR detection [email protected] from 64.26 to 81.46 compared with the single-USV baseline. RGB features further improve intermediate f
Open source S4
usion, increasing [email protected] by up to 4.55 percentage points. LiDAR and RGB transfer experiments also show the value of the generated data as a pretraining source for real maritime detection. The project are available at https://github.com/lizaowo/MMUSV-Sim
Open source S5
Implications for data suppliers
RoboSignal interpretation and collection questions, not statements of buyer demand.
- Confirm the required data type and collection setting with the buyer; this source does not establish a complete collection specification.
- Compare the reported units and scope before using these quantities in a budget. Recording hours, sensor-hours and trajectories are different measures.
- Check the exact dataset version, license, modalities and task coverage against the intended use.
- Validate demand and acceptance criteria with a buyer before scaling. Publication, popularity and a research result do not establish a purchase commitment.
Source：arXiv Robotics — research abstracts · arxiv.org