Driver Behavior Prediction Based on Environmental Observation Using Fuzzy Hidden Markov Model Alif Rizqullah Mahdi, Yul Yunazwin Nazaruddin, Miranti Indar Mandasari
Institut Teknologi Bandung, National Center for Sustainable Transportation Technology
Abstract
The development of autonomous vehicle system has progressed rapidly in recent years. One challenge that persists is the capability of the autonomous system in responding to human drivers. Human behavior is an integral part of driving, changing lanes and speed adjustments are all determined by the driver behavior. However, human behavior is unpredictable and immeasurable. Some traffic accidents are caused due to the erratic behavior of the driver. Although, traffic laws such as in Indonesia, regulates the use of lanes with regards to the speed of the vehicle. The behavior of drivers on lane are more likely to be influenced by the regulation. This paper proposes a novel method of predicting the behavior of drivers by utilizing the concept of fuzzy Hidden Markov Model (fuzzy HMM). HMM has been proven to be reliable in predicting human behavior by observing measurable states to determine unmeasurable hidden states. The use of fuzzy logic is to mimic the way that humans perceive speeds of other vehicles. The fuzzy logic determines the relative observed state of other vehicles according to the measured velocity of an ego vehicle and the observed state of observed vehicles. Observation data is obtained by equipping an ego vehicle with an action camera. The observed data, in the form of a video, is then discretized for every 2 seconds. The resulting sequence of images are processed to determine several variables: speed and state of the observed vehicles (lane position and speed), and the time instance of the observation. The fuzzy HMM is generated based on the observational data. A predictor created using fuzzy HMM equipped with a training and prediction algorithm is successful in predicting the behavior of other drivers on the road.
