Driver Behavior Prediction Based on Environmental Observation Using Fuzzy Hidden Markov Model

The development of autonomous vehicle systems has progressed rapidly in recent years. One challenge that persists is the capability system to respond human drivers. Human behavior an integral part driving; thus, driver determines changing lanes and speed adjustments. However, unpredictable immeasurable. Some traffic accidents are caused due erratic driver. Although, laws, such as Indonesia, regulate use concerning vehicle’s speed. drivers’ lane more likely be influenced by regulation. This paper proposes a novel method predicting utilizing concept fuzzy Hidden Markov Model (fuzzy HMM). HMM been proven reliable observing measurable states determine unmeasurable hidden states. logic mimic way humans perceive speeds other vehicles. relative observed state vehicles according measured velocity ego Observation data obtained equipping with action camera. data, form video, then discretized every 2 seconds. resulting sequence images processed several variables: (lane position speed) time instance observation. generated based on observational data. A predictor created using equipped training prediction algorithm successfully predicts drivers road.