A Study on Artificial Intelligent-based Trajectory Pattern Mining for Smart Vehicular Traffic
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Recently, with the rapid growth and widespread use of GPS devices, RFID sensors using Internet of Things, wireless communication, and artificial intelligent, it has become possible to track various moving objects in real-world. In addition, an increasing number of trajectories are being collected and stored in databases, often containing valuable knowledge that requires urgent analysis. As the adoption of location acquisition technologies increases, here is a growing opportunity to collect large-scale spatio-temporal datasets and discover actionable knowledge about moving behaviors, fostering the development of new applications and services. The main functionalities and modules of activity pattern discovery system include trajectory preprocessing, trajectory clustering, periodic pattern discovery, and an extensible visualization module for parameter interaction and result presentation. In the experimental part, we validate the accuracy and efficiency of activity pattern discovery system using taxi trajectory data from Korea. In this paper, we develop an extension of the sequential pattern mining paradigm to analyze trajectories of moving objects in this direction. In addition, we introduce trajectory patterns as concise descriptions of frequent behaviors in terms of space and time. In this setting, we propose a general formalization for the novel mining problem and then study several different instances of varying complexities. Fianlly, we empirically evaluate different approaches on real-world data and synthetic benchmarks to compare their strengths and weaknesses.
Abstract
Recently, with the rapid growth and widespread use of GPS devices, RFID sensors using Internet of Things, wireless communication, and artificial intelligent, it has become possible to track various moving objects in real-world. In addition, an increasing number of trajectories are being collected and stored in databases, often containing valuable knowledge that requires urgent analysis. As the adoption of location acquisition technologies increases, here is a growing opportunity to collect large-scale spatio-temporal datasets and discover actionable knowledge about moving behaviors, fostering the development of new applications and services. The main functionalities and modules of activity pattern discovery system include trajectory preprocessing, trajectory clustering, periodic pattern discovery, and an extensible visualization module for parameter interaction and result presentation. In the experimental part, we validate the accuracy and efficiency of activity pattern discovery system using taxi trajectory data from Korea. In this paper, we develop an extension of the sequential pattern mining paradigm to analyze trajectories of moving objects in this direction. In addition, we introduce trajectory patterns as concise descriptions of frequent behaviors in terms of space and time. In this setting, we propose a general formalization for the novel mining problem and then study several different instances of varying complexities. Fianlly, we empirically evaluate different approaches on real-world data and synthetic benchmarks to compare their strengths and weaknesses.
참고문헌 (15)
- 채예은, "SUMO를 이용한 VANET에서의 애드 혹 라우팅 프로토콜의 성능 비교", 한국정보기술학회논문지, 15(2), pp.91~98 (2017), 한국정보기술학회[DOI]
- 서봉수, "Energy-Efficient Data Delivery Strategy for V2I Systems with Power-Constrained Road Side Units", 한국정보기술학회논문지, 19(5), pp.101~109 (2021), 한국정보기술학회[DOI]
- M. N. Tahir, "Analysis of VANET wireless networking technologies in realistic environments", Proceedings of 2021 IEEE Radio and Wireless Symposium (RWS), pp.17~22 (2021)
- 서봉수, "V2I 시스템에서 가변 전송 거리에 기반한 근사 최적의 전달 지연을 제공하는 데이터 전달 방법", 한국정보기술학회논문지, 17(12), pp.93~100 (2019), 한국정보기술학회[DOI]
- 김시관, "차량용 네트워크에서의 성능 향상을 위한 라우팅 프로토콜", 한국정보기술학회논문지, 19(7), pp.67~73 (2021), 한국정보기술학회[DOI]
- A. Keränen, "The one simulator for DTN protocol evaluation", SIMUTools’09: 2nd International Conference on Simulation Tools and Techniques (2009)
- T. Spyropoulos, "Spray and wait: An efficient routing scheme for intermittently connected mobile networks", Proceedings of ACM SIGCOMM 2005 Workshop on Delay Tolerant Networking and Related Networks (WDTN-05), pp.252~259 (2005)
- A. Silva, "Improvement and performance evaluation of GPSR-based routing techniques for vehicular ad hoc networks", IEEE Access, 7, pp.21722~21733 (2019)
- R. S. Sutton, "Policy gradient methods for reinforcement learning with function approximation", Advances in neural information processing systems, pp.1057~1063 (2000)
- J. Yuan, "Driving with knowledge from the physical world", Proceedings of the 17th ACM SIGKDD international conference on knowledge discovery and datamining, pp.316~324 (2011)
- F. Verhein, "Mining complex spatio-temporal sequence patterns", Proceedings of the 2009 SIAM international conference on datamining, Sparks, pp.605~616 (2009)
- W. Han, "Aparallel online trajectory compression approach for supporting big data workflow", Computing 2018, 100(1), pp.3~20 (2018)
- 안철범, "자가 적응형 예측 서비스를 위한 스마트 차량 네트워크 설계", 한국지식정보기술학회 논문지, 15(6), pp.1065~1073 (2020), 한국지식정보기술학회[DOI]
- 김완태, "AI 알고리즘을 이용한 스마트 차량의 위협도와 우선순위 결정 방법", 한국지식정보기술학회 논문지, 18(3), pp.755~762 (2023), 한국지식정보기술학회[DOI]
- Symphorien Karl Yoki Donzia, "Design and Implementation of Autonomous Driving Cars Based on Adaptive Autoware Application", 한국지식정보기술학회 논문지, 16(5), pp.1003~1012 (2021), 한국지식정보기술학회[DOI]