Autonomous Vessel Test Data for Detecting Weak Emergent Behavior

Experimental input and output data from an industrial case study on an Autonomous Surface Vessel performing a Search and Rescue mission. The data was generated using the Taguchi method for Design of Experiments and analyzed with regression techniques by researchers at the University of South-Eastern Norway. The dataset is organized across 10 spreadsheet sheets covering screening, investigation, and analysis phases.

Use Cases

• Training regression models to predict system performance based on experimental parameters described in the study
• Benchmarking methods for detecting weak emergent behavior in engineered complex systems
• Comparing the effectiveness of Taguchi orthogonal arrays versus fractional factorial designs for parameter space exploration
• Analyzing the relationship between noise parameters and system failure in autonomous maritime operations

Strengths

• Dataset is based on a real industrial case study conducted in a company-specific test arena, providing practical relevance.
• The experimental design follows the structured Taguchi method and includes multiple investigation settings for systematic analysis.
• Data organization across 10 distinct sheets suggests a detailed breakdown of the experimental process and results.

Limitations

• Row count, column names, and sample data are unknown, making it difficult to assess the dataset's scale and structure prior to download.
• The description lacks specific details on file formats, size, and last update date, limiting metadata completeness.
• Column-level documentation is absent; field semantics must be inferred after download.

Provenance

Source

University of South-Eastern Norway

Collection Method

Experiments were set up using the Taguchi method and conducted in a case company's test arena, with data generated from an Autonomous Surface Vessel on a Search and Rescue mission.