This paper presents the use of a video camera for predictingvehicle trajectory. It analyses the data available from the videocamera, the features which can be used for trajectory prediction and howthese relate to the real world. A recursive least squares basedestimator is designed which is robust with respect to horizon problemsand deals with distance varying measurement noise. The resultsillustrate the estimation and comprises simulated data, to illustraterobustness, and video data to show typical road features
展开▼