Car-following Behavior Model Learning Using Timed Automata
نویسندگان
چکیده
Learning driving behavior is fundamental for autonomous vehicles to “understand” traffic situations. This paper proposes a novel method for learning a behavioral model of carfollowing using automata learning algorithms. The model is interpretable for car-following behavior analysis. Frequent common state sequences are extracted from the model and clustered as driving patterns. The Next Generation SIMulation dataset on the I-80 highway is used for learning and evaluating. The experimental results demonstrate high accuracy of car-following model fitting.
منابع مشابه
A New Model of Car Following Behavior Based on Lane Change Effects Using Anticipation and Evaluation Idea
This paper aims to investigate a new and intricate behavior of immediate follower during the lane change of leader vehicle. Accordingly, the mentioned situation is a transient state in car following behavior during which the follower vehicle considerably deviates from conventional car following models for a limited time, which is a complex state including lateral and longitudinal movement simul...
متن کاملUsing Interval Petri Nets and Timed Automata for Diagnosis of Discrete Event Systems (DES)
A discrete event system (DES) is a dynamic system that evolves in accordance with the abrupt occurrence, at possibly unknown irregular intervals, of physical events. Because of the special nature of these systems, different tools are currently used for their analysis, design and modeling. The main focus of this paper is the presentation of a new modeling approach of Discrete Event Systems. The ...
متن کاملModel Predictive Control System Design using ARMAX Identification Method for Car-following Behavior
The control of car following is essential due to its safety and its operational efficiency. For this purpose, this paper builds a model of car following behavior based on ARMAX structure from a real traffic dataset and design a Model Predictive Control (MPC) system. Based on the relative distance and relative acceleration of each instant, the MPC predicts the future behavior of the leader vehic...
متن کاملPrediction of Car Following Behavior Based on the Instantaneous Reaction Time using an ANFIS-CART Based Model
Car-following models are among the most important components of micro traffic flow simulation which is studiedby transportation experts to evaluate new applications of intelligent transportation systems. Until now, several carfollowingmodels have been proposed. An obvious disadvantage of the former models is the great number of parameterswhich are difficult to calibrate. In th...
متن کاملA likelihood-ratio test for identifying probabilistic deterministic real-time automata from positive data1
Timed automata (TAs) are finite state models that represent timed events using an explicit notion of time, i.e., using numbers. They can be used to model and reason about real-time systems. In these system, each occurrence of a symbol (event) is associated with a time value, i.e., its time of occurrence. TAs can be used to accept or generate a sequence (a1, t1)(a2, t2)(a3, t3) . . . (an, tn) of...
متن کامل