Dataset Information

Online legal driving behavior monitoring for self-driving vehicles.

ABSTRACT: Defined traffic laws must be respected by all vehicles when driving on the road, including self-driving vehicles without human drivers. Nevertheless, the ambiguity of human-oriented traffic laws, particularly compliance thresholds, poses a significant challenge to the implementation of regulations on self-driving vehicles, especially in detecting illegal driving behaviors. To address these challenges, here we present a trigger-based hierarchical online monitor for self-assessment of driving behavior, which aims to improve the rationality and real-time performance of the monitoring results. Furthermore, the general principle to determine the ambiguous compliance threshold based on real driving behaviors is proposed, and the specific outcomes and sensitivity of the compliance threshold selection are analyzed. In this work, the effectiveness and real-time capability of the online monitor were verified using both Chinese human driving behavior datasets and real vehicle field tests, indicating the potential for implementing regulations in self-driving vehicles for online monitoring.

SUBMITTER: Yu W

PROVIDER: S-EPMC10776857 | biostudies-literature | 2024 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Online legal driving behavior monitoring for self-driving vehicles.

Yu Wenhao W Zhao Chengxiang C Wang Hong H Liu Jiaxin J Ma Xiaohan X Yang Yingkai Y Li Jun J Wang Weida W Hu Xiaosong X Zhao Ding D

Nature communications 20240109 1

Defined traffic laws must be respected by all vehicles when driving on the road, including self-driving vehicles without human drivers. Nevertheless, the ambiguity of human-oriented traffic laws, particularly compliance thresholds, poses a significant challenge to the implementation of regulations on self-driving vehicles, especially in detecting illegal driving behaviors. To address these challenges, here we present a trigger-based hierarchical online monitor for self-assessment of driving beha ...[more]

PMID: 38195672

Similar Datasets

Project description:Background: This study investigated the determinants (with a special emphasis on the role of diabetes app use, use of continuous glucose monitoring (CGM) device, and self-care behavior) of glycemic control of type 1 and type 2 diabetes mellitus (DM). Methods: A web-based survey was conducted using diabetes Facebook groups, online patient-forums, and targeted Facebook advertisements (ads). Demographic, CGM, diabetes app use, and self-care behavior data were collected. Glycemic level data were categorized into hyperglycemia, hypoglycemia, and good control. Multinomial logistic regression stratified by diabetes type was performed. Results: The survey URL was posted in 78 Facebook groups and eight online forums, and ten targeted Facebook ads were conducted yielding 1854 responses. Of those owning smartphones (n = 1753, 95%), 1052 (62.6%) had type 1 and 630 (37.4%) had type 2 DM. More than half of the type 1 respondents (n = 549, 52.2%) and one third the respondents with type 2 DM (n = 210, 33.3%) reported using diabetes apps. Increased odds of experiencing hyperglycemia were noted in persons with type 1 DM with lower educational status (Adjusted Odds Ratio (AOR) = 1.7; 95% Confidence Interval (CI): 1.21⁻2.39); smokers (1.63, 95% CI: 1.15⁻2.32), and high diabetes self-management concern (AOR = 2.09, 95% CI: 1.15⁻2.32). CGM use (AOR = 0.66, 95% CI: 0.44⁻1.00); "general diet" (AOR = 0.86, 95% CI: 0.79⁻0.94); and "blood glucose monitoring" (AOR = 0.88, 95%CI: 0.80⁻0.97) self-care behavior reduced the odds of experiencing hyperglycemia. Hypoglycemia in type 1 DM was reduced by using CGM (AOR = 0.24, 95% CI: 0.09⁻0.60), while it was increased by experiencing a high diabetes self-management concern (AOR = 1.94, 95% CI: 1.04⁻3.61). Hyperglycemia in type 2 DM was increased by age (OR = 1.02, 95% CI: 1.00⁻1.04); high self-management concern (AOR = 2.59, 95% CI: 1.74⁻3.84); and poor confidence in self-management capacity (AOR = 3.22, 2.07⁻5.00). Conversely, diabetes app use (AOR = 0.63, 95% CI: 0.41⁻0.96) and "general diet" self-care (AOR = 0.84, 95% CI: 0.75⁻0.94), were significantly associated with the reduced odds of hyperglycemia. Conclusion: Diabetes apps, CGM, and educational interventions aimed at reducing self-management concerns and enhancing dietary self-care behavior and self-management confidence may help patients with diabetes to improve glycemic control.

Dataset Information

Online legal driving behavior monitoring for self-driving vehicles.

Publications

Online legal driving behavior monitoring for self-driving vehicles.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets