Journal
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
Volume 22, Issue 6, Pages 3654-3663Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TITS.2020.3037902
Keywords
Sensors; Navigation; Mathematical model; Reliability; Vehicle dynamics; Connected vehicles; Training; Ant colony optimization; intelligent transportation systems; recurrent learning; service dissemination; time-dependent dissemination
Ask authors/readers for more resources
The article introduces a method called the ant-inspired recurrent learning model (ARLM) aimed at maximizing the navigation service response rate of vehicles through forward and backward ant agents and recurrent learning, and reducing sensing and response time through conditional verification.
Intelligent Transportation System (ITS) serves as the on-the wheel communication and service platform for the real-world driving users. Navigation service and traffic information flow among the connected vehicles relies on the available resources and infrastructure units. Appropriate sensing and selection of infrastructure units for seamless navigation responses and information flow in the dynamic environment is facilitated using bio-inspired learning in this article. This method named as ant-inspired recurrent learning model (ARLM) introduced in this article is focused to improve the sensing and response rate of the navigation-based services for the vehicles. This model relies on forward and backward ant agents and recurrent learning for maximizing the navigation service response rate of the vehicles. In this model, the training sets are differentiated on the basis of connection probability and learning depreciation to retain the service rate through different learning iterates. The conditional verification for sensing and retaining active link helps to reduce the sensing and response time irrespective of the varying vehicle and request densities. The performance of the proposed model is verified using suitable experimental analysis and the metrics information flow rate, service sensing, sensing time, response rate, and response time are analyzed for assessing ARLM.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available