Model-Based Design of Integrated Transportation Systems Using Approximate Dynamic Programming.

2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC)(2023)

引用 0|浏览2
暂无评分
摘要
Based on recent developments and technological advances, the complexity and number of systems and services for intelligent transportation systems (ITS) has risen dramatically. The interactions between these systems and services require systematic development methods for their integrated structural design as well as their behavioral planning in terms of integrated control strategies, which can address multiple as well as heterogenous goals. To address this situation, in this work, we propose an approach for integrated modeling of heterogenous ITS systems and their integrated evaluation using approximate dynamic programming (ADP). For this, we introduce a methodology which incorporates the necessary design principles and requirements to address above challenges. Then, we present an integrated systems modeling technique which is used for problem formulation. In addition to problem formulation, it provides a generalist problem solution framework using ADP, which can be applied to multi-domain and heterogenous model problems. To demonstrate domain-specific application of our approach and systems modeling technique, we provide an exemplary model problem and its solution for an autonomous mobility on demand system in a demand responsive transport (DRT) scenario.
更多
查看译文
关键词
Transport System,Dynamic Programming,Approximate Dynamic Programming,Model System,Control Strategy,Rational Design,Integrated Model,Number Of Systems,Heterogeneous Systems,Heterogeneous Model,Intelligent Transportation Systems,Planned Behavior,Value Function,Behavioral Model,Model Evaluation,State Space,Power System,Dynamic Properties,Design Problem,Static Properties,Value Function Approximation,Emergent Properties,Proximal Policy Optimization,Vehicle Acceleration,Transition Function,Autonomous Vehicles,Discrete Action,Configuration Space,Start Position
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要