Dataset Information

Building causal models for finding actual causes of unmanned aerial vehicle failures.

ABSTRACT: Finding actual causes of unmanned aerial vehicle (UAV) failures can be split into two main tasks: building causal models and performing actual causality analysis (ACA) over them. While there are available solutions in the literature to perform ACA, building comprehensive causal models is still an open problem. The expensive and time-consuming process of building such models, typically performed manually by domain experts, has hindered the widespread application of causality-based diagnosis solutions in practice. This study proposes a methodology based on natural language processing for automating causal model generation for UAVs. After collecting textual data from online resources, causal keywords are identified in sentences. Next, cause-effect phrases are extracted from sentences based on predefined dependency rules between tokens. Finally, the extracted cause-effect pairs are merged to form a causal graph, which we then use for ACA. To demonstrate the applicability of our framework, we scrape online text resources of Ardupilot, an open-source UAV controller software. Our evaluations using real flight logs show that the generated graphs can successfully be used to find the actual causes of unwanted events. Moreover, our hybrid cause-effect extraction module performs better than a purely deep-learning based tool (i.e., CiRA) by 32% in precision and 25% in recall in our Ardupilot use case.

SUBMITTER: Zibaei E

PROVIDER: S-EPMC10880731 | biostudies-literature | 2024

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Building causal models for finding actual causes of unmanned aerial vehicle failures.

Zibaei Ehsan E Borth Robin R

Frontiers in robotics and AI 20240207

Finding actual causes of unmanned aerial vehicle (UAV) failures can be split into two main tasks: building causal models and performing actual causality analysis (ACA) over them. While there are available solutions in the literature to perform ACA, building comprehensive causal models is still an open problem. The expensive and time-consuming process of building such models, typically performed manually by domain experts, has hindered the widespread application of causality-based diagnosis solut ...[more]

PMID: 38384357

Similar Datasets

Project description:BackgroundThe technology of cotton defoliation is essential for mechanical cotton harvesting. Agricultural unmanned aerial vehicle (UAV) spraying has the advantages of low cost, high efficiency and no mechanical damage to cotton and has been favored and widely used by cotton planters in China. However, there are also some problems of low cotton defoliation rates and high impurity rates caused by unclear spraying amounts of cotton defoliants. The chemical rate recommendation and application should be based upon crop canopy volume rather than on land area. Plant height and leaf area index (LAI) is directly connected to plant canopy structure. Accurate dynamic monitoring of plant height and LAI provides important information for evaluating cotton growth and production. The traditional method to obtain plant height and LAI was s a time-consuming and labor-intensive task. It is very difficult and unrealistic to use the traditional measurement method to make the temporal and spatial variation map of plant height and LAI of large cotton fields. With the application of UAV in agriculture, remote sensing by UAV is currently regarded as an effective technology for monitoring and estimating plant height and LAI.ResultsIn this paper, we used UAV RGB photos to build dense point clouds to estimate cotton plant height and LAI following cotton defoliant spraying. The results indicate that the proposed method was able to dynamically monitor the changes in the LAI of cotton at different times. At 3 days after defoliant spraying, the correlation between the plant height estimated based on the constructed dense point cloud and the measured plant height was strong, with [Formula: see text] and RMSE values of 0.962 and 0.913, respectively. At 10 days after defoliant spraying, the correlation became weaker over time, with [Formula: see text] and RMSE values of 0.018 and 0.027, respectively. Comparing the actual manually measured LAI with the estimated LAI based on the dense point cloud, the [Formula: see text] and RMSE were 0.872 and 0.814 and 0.132 and 0.173 at 3 and 10 days after defoliant spraying, respectively.ConclusionsDense point cloud construction based on UAV remote sensing is a potential alternative to plant height and LAI estimation. The accuracy of LAI estimation can be improved by considering both plant height and planting density.

Dataset Information

Building causal models for finding actual causes of unmanned aerial vehicle failures.

Publications

Building causal models for finding actual causes of unmanned aerial vehicle failures.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets