PREVIOUS FRAME SENSE STATE USED FOR CURRENT FRAME PREDICTION

A system for navigating a host vehicle relative to a road segment may receive object detection information relating to detection of one or more objects represented in a first image frame captured by at least one camera onboard the host vehicle; receive a second image frame captured by the at least one camera onboard the host vehicle; receive a portion of a map representative of the road segment; provide to a trained model the object detection information, the second image frame, and a representation of at least a portion of the at least one drivable path, wherein the trained model is configured to identify at least one target vehicle represented in the second image frame based, at least in part, on inferences relative to the object detection information, the second image frame, and the representation of at least a portion of the at least one drivable path.