Passenger intent prediction for automatic car entry system

Abstract

A vehicle access control system for predicting a user's intent to enter a vehicle is achieved through an analysis of video data captured by a plurality of external cameras mounted on the vehicle. The plurality of cameras capture the video data as the user approaches the vehicle. The system includes a processor with an intent prediction module. A memory stores instructions which, when executed by the processor, enable the processor to analyze visual cues of the user from the video data, including body posture, gaze direction, and proximity to the vehicle. Furthermore, the processor extracts user data including a trajectory, head orientation, and body orientation of the user from the video data. The intent prediction module, using the user data, predicts the intent of the user to enter the vehicle and triggers a corresponding vehicle entry action.