Trajectory generation using road network model

A method and system of trajectory generation using a road network model comprising; obtaining, using at least one processor of a vehicle, a location of the vehicle, sensor data collected at the location; obtaining and map data for the location (701). Generating, using one or more processors, at least one possible trajectory for at least one object at the location, wherein the possible trajectory is constrained in accordance with the map data (702). Predicting, using a machine learning model, a score for the at least one trajectory (703). The system may generate feature vectors generated from an image embedding of the sensor data and the map data. Point clouds may be utilised in conjunction with LiDAR.